41 Commits

Author	SHA1	Message	Date
Sameer Kankute	cfcdf8714a	feat: litellm oss 110626 (#30202) ... * Add gpt-realtime-whisper Realtime transcription support (OpenAI + Azure) (#29775) * Add gpt-realtime-whisper Realtime transcription support (OpenAI + Azure) Adds first-class support for the gpt-realtime-whisper streaming speech-to-text model, which uses the Realtime transcription session API rather than the file-based /audio/transcriptions path. Model registration: registers gpt-realtime-whisper and azure/gpt-realtime-whisper with audio-duration pricing (input_cost_per_second = 0.017/60, matching the published $0.017/minute input audio rate). REST endpoint: implements POST /v1/realtime/transcription_sessions (plus /realtime and /openai/v1 aliases) to mint an ephemeral transcription session for the WebRTC flow. Adds request/response types, OpenAI and Azure URL builders, a shared base handler (refactored from the client_secrets handler), the acreate_realtime_transcription_session SDK function, and route registration. The proxy encrypts the ephemeral key returned under client_secret.value and records the session type in the token so the follow-up /realtime/calls replays type=transcription rather than type=realtime. WebSocket: forwards intent=transcription through to the Azure handler (OpenAI already received it) with URL-encoding, so gpt-realtime-whisper opens a transcription session. Transcription-only sessions no longer trigger an erroneous response.create. Cost tracking: transcription sessions emit no response.done events; their usage arrives on conversation.item.input_audio_transcription.completed as {type: duration, seconds}. That usage is captured out-of-band (usage only, no transcript duplication) and billed by input_cost_per_second, with a token-billed fallback for token-priced transcription models. Adds tests for pricing math, URL builders, request/response types, the proxy route and SDK function, WebSocket intent forwarding, transcription-session streaming behavior, and the /realtime/calls session-type replay. * Address PR review: URL-encode all Azure WS query params; forward query_params through provider_config branch * Address PR review: session_type validation, model auth fix, cost perf, billing fallback, detail/docs cleanup * Improve test coverage: detection from backend, error paths, unknown usage type, resolved_model None * Backport realtime transcription websocket fixes * Enforce authorized realtime transcription model * Enforce realtime transcription model access * Enforce realtime resolved model scopes * Enforce WebRTC transcription model scope * Lazy evaluate debug log in pass-through endpoint (#30177) * Pass through debug lazy logging * fix(proxy): convert remaining eager pass-through debug logs to lazy formatting * fix(parallel_ai): migrate search integration from v1beta to v1 endpoint (#30157) * fix(parallel_ai): migrate search integration from v1beta to v1 endpoint The Parallel Search API moved from /v1beta/search (processor: base/pro, parallel-beta header) to /v1/search (mode: turbo/basic/advanced, no beta header). Request fields moved too: max_results, source_policy, and excerpt settings are now nested under advanced_settings, and source_policy uses include_domains/exclude_domains. The v1 response returns publish_date per result, which now maps to SearchResult.date instead of being hardcoded to None. The legacy processor param is mapped to the equivalent mode so existing callers keep working. * fix(parallel_ai): default mode to basic and simplify param handling The v1 API defaults to advanced mode when mode is omitted, while v1beta defaulted to the base processor. Without an explicit default, callers who pass no mode would be silently upgraded to a tier costing 2.25x more while litellm's cost map reports the basic-tier price. Sending mode=basic preserves the v1beta default and keeps cost tracking accurate. Also replaces the handled_params set with pop-as-consumed param handling so mapped params no longer need to be tracked in two places, and extends the tests to pin the default mode, processor=base mapping, mode-over-processor precedence, and top-level v1 param passthrough. * fix(parallel_ai): avoid double /v1 when api_base is already versioned A PARALLEL_AI_API_BASE like https://api.parallel.ai/v1 previously produced .../v1/v1/search. Strip a trailing /v1 before appending the search path and cover the api_base variants with a parametrized test. --------- Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai> * feat(focus): add Mavvrik destination for FOCUS export (#29935) * fix: preserve responses streaming flag (#30189) * fix: preserve responses streaming flag * test: cover async responses streaming flag * fix(spend/daily-activity): stable offset pagination via id tiebreaker (#30164) (#30167) date alone is not a unique sort key for LiteLLM_DailyUserSpend or LiteLLM_DailyTeamSpend (many rows per date: api_key x model x model_group x provider x endpoint). Offset pagination over a non-unique sort landed on arbitrary boundaries, so a client paging through all results and summing per-page metrics (the Usage dashboard) got non-deterministic totals - sometimes inflated, sometimes deflated, different at different page_size values. Adding the row's UUID id (present on both tables) as a secondary sort gives every page a stable cursor. order=[{date desc}, {id asc}]. Fixes #30164 * fix(oci): inject a default maxTokens so omitted max_tokens doesn't truncate responses (#30018) * fix(oci): inject default maxTokens so omitted max_tokens doesn't truncate OCI GenAI applies a tiny server-side maxTokens default (~20 tokens) when the request omits it, so any call that doesn't send max_tokens comes back cut off mid-string with finishReason "length". MLflow judges never send max_tokens, so their JSON responses arrived as unterminated strings and json.loads failed in MLflow's gateway adapter. When no maxTokens/maxCompletionTokens target is set, inject DEFAULT_OCI_CHAT_MAX_TOKENS (env-overridable, defaults 4096), mirroring the Anthropic config's default-max-tokens behaviour. An explicit max_tokens still wins, and reasoning models still route to maxCompletionTokens. Used a fixed default rather than the catalog max_output_tokens because the catalog value is unreliable for some models (grok-4 reports max_output_tokens equal to its context window, not a real output cap, which would risk 400s). Adds TestOCIDefaultMaxTokens covering Cohere and generic injection, the explicit-override case, and the reasoning maxCompletionTokens branch. * test(oci): e2e regression that omitted max_tokens isn't truncated Real-proxy integration test asserting a chat completion that omits max_tokens completes with finish_reason "stop" instead of being cut off at OCI's ~20-token server default. Fails before the maxTokens-default injection (finish_reason "length", ~19 tokens), passes after. * test(oci): update cohere default-params test for injected maxTokens test_cohere_default_parameters asserted no maxTokens was injected, encoding the old behaviour where OCI's ~20-token server default truncated responses. Now that transform_request injects DEFAULT_OCI_CHAT_MAX_TOKENS, assert maxTokens equals that default while the other params (topK/topP/frequencyPenalty) stay pass-through with no hardcoded default. * fix(oci): make DEFAULT_OCI_CHAT_MAX_TOKENS a plain constant Drop the os.getenv override. The env knob was not requested and introducing a new env var forced a cross-repo dependency on litellm-docs (test_env_keys.py validates every referenced env var against the docs table there). A plain 4096 constant keeps the PR self-contained; callers who want a different limit pass max_tokens explicitly per request. * fix(oci): route all OpenAI commercial models to maxCompletionTokens OCI serves OpenAI models (gpt-4.1, gpt-5.1 through 5.5, o-series) that the litellm catalog doesn't track, so the supports_reasoning lookup returned False for them and the provider sent maxTokens, which the reasoning families reject with HTTP 400. With the injected default maxTokens this broke every request to those models, not just ones with an explicit max_tokens. Route the whole openai.* vendor prefix to maxCompletionTokens since OpenAI accepts max_completion_tokens on every chat model; the openai.gpt-oss-* open weights are served by OCI's own stack and keep maxTokens. Verified live against gpt-5.2, gpt-5, gpt-4o, gpt-4.1, gpt-oss-120b, llama-3.3, command-a and grok-3-mini * test(oci): hoist transformation imports and drop unused ones Makes the generic-chat test file ruff-clean: the per-test local imports of OCIChatConfig/OCIVendors shadowed the module-level import (F811) and left it unused (F401), and json plus three OCI type imports were never referenced * fix(oci): translate response_format json_schema to OCI's accepted shape (#29691) * fix(oci): translate response_format json_schema to OCI's accepted shape OCI GenAI rejected every json_schema response_format with HTTP 400 "Please pass in correct format of request", which broke structured-output callers such as MLflow LLM judges (they always send a json_schema). The provider forwarded OpenAI's raw json_schema body unchanged. For GENERIC models OCI's ResponseJsonSchema accepts only name/description/schema/isStrict, so OpenAI's `strict` key (and any other extra) 400s the request; the key must be renamed to isStrict and the body whitelisted. For Cohere models there is no JSON_SCHEMA type at all; the schema has to ride on JSON_OBJECT as {"type": "JSON_OBJECT", "schema": ...}. Cohere type values must also be the canonical uppercase TEXT/JSON_OBJECT. _normalize_response_format now branches by vendor and emits the exact shape each one accepts (verified live against OCI GenAI for Cohere, Meta, Gemini and Grok). Drops the unused, incorrect Cohere response-format pydantic models. Two existing tests asserted the broken behavior (lowercase type, raw jsonSchema on Cohere); they are rewritten to assert the corrected shape, and generic/Cohere json_schema regression tests are added. * fix(oci): raise early on json_schema response_format with no body A GENERIC model request with {"type": "json_schema"} and no json_schema object fell through to the JSON_OBJECT branch and emitted a bodyless {"type": "JSON_SCHEMA"}, which OCI rejects with an opaque HTTP 400. Raise a descriptive 400 at translation time instead. Cohere is unaffected since it always maps to JSON_OBJECT. * test(oci): gateway integration test for response_format json_schema Added to tests/integration/ (the real-network integration suite) reusing the existing OCI proxy harness, not tests/llm_translation/ which is mock-only. --------- Co-authored-by: Sameer Kankute <sameer@berri.ai> * fix(oci): accept default n=1 on Cohere instead of hard-failing (#29705) * fix(oci): accept default n=1 on Cohere instead of hard-failing Cohere on OCI has no numGenerations field, so n was mapped to False and map_openai_params raised "param `n` is not supported on OCI" whenever a client sent n. But n=1 (and None) is the OpenAI default single-generation request, which every OCI model produces anyway, so standard clients that always send n=1 (such as the MLflow gateway) were rejected with a 500. Drop n=1/None silently for Cohere; only n>1 is genuinely unsupported and still raises (or drops under drop_params). Generic models are unaffected and keep numGenerations, including n>1. * docs(oci): explain why n is not advertised for Cohere despite tolerating n=1 * test(oci): gateway integration test for Cohere default n=1 Added to tests/integration/ (the real-network integration suite) reusing the existing OCI proxy harness, not tests/llm_translation/ which is mock-only. --------- Co-authored-by: Sameer Kankute <sameer@berri.ai> * fix(oci): drop max_retries instead of hard-failing on OCI (#29727) max_retries is a litellm-level control param (litellm applies retries itself), not a generation param OCI accepts. The provider mapped it to False and raised "param `max_retries` is not supported on OCI" whenever it was present. The litellm proxy injects max_retries on every request, so any OCI call through the proxy 500'd unless drop_params was set. Drop max_retries silently in map_openai_params. Adds a unit test (Cohere and generic) and a gateway integration test that a plain request succeeds through a proxy without drop_params. Co-authored-by: Sameer Kankute <sameer@berri.ai> * fix(spend-logs): rehydrate metadata JSONB text on ui_view_spend_logs (#29682) Fixes #29674. `/spend/logs/ui` raw-SQL path returns the JSONB metadata column as a string — prisma's query_raw skips the ORM-layer hydration. The UI reads metadata.status / metadata.error_information as object fields, so provider-failure rows look like successes. Fix: json.loads the metadata field right after query_raw, fall back to {} on malformed JSON. 3 existing error-code/error-message tests called json.loads on response.data[0]["metadata"] — they were leaning on the bug. Updated to read the dict directly. Plus 2 new regression tests (failure metadata roundtrip + invalid-json fallback). Reverting the fix makes both new tests fail with AssertionError: metadata should be dict, got <class 'str'>. * fix(proxy): release max_parallel_requests slot when a stream is cancelled mid-flight (#27955) (#30020) * fix(proxy): release max_parallel_requests slot when a stream is cancelled mid-flight (#27955) * fix: refund max_parallel_requests on disconnect from outer streaming generators The cancellation refund previously lived in async_post_call_streaming_iterator_hook, but that hook is nested inside the outer streaming generators and a nested async generator only receives GeneratorExit on garbage collection (non-deterministic). With only the v3 limiter enabled, /chat/completions also bypasses the hook entirely (needs_iterator_wrap() is false). Move the release into async_data_generator and async_streaming_data_generator, the generators Starlette closes on client disconnect, so the refund fires deterministically on every streaming route. Warn when no event loop is running, and document the window TTL refresh on the decrement * fix(mcp): propagate model into model_call_details for passthrough tool calls (#30122) * fix(mcp): propagate model into model_call_details for passthrough tool calls The @client decorator on call_mcp_tool creates the logging object via function_setup without a model kwarg, so model_call_details["model"] starts as None. execute_mcp_tool only set logging_obj.model as an instance attribute, which the spend-log writer never reads (it reads kwargs["model"] from model_call_details). MCP passthrough tools/call rows therefore persisted with model="" while list_tools rows showed "MCP: list_tools", degrading the Logs UI display and bucketing all MCP tool spend under an empty model in DailyUserSpend. Propagate the model into model_call_details alongside the existing attribute assignment so the StandardLoggingPayload and SpendLogs writer pick it up. Covers the /mcp passthrough, REST /mcp-rest/tools/call, and orchestrated paths (the latter already passed model into function_setup, so this is a no-op there). * test(mcp): trim regression test docstring * fix(mcp): surface upstream challenges for delegated OAuth (#30124) * fix(mcp): surface upstream challenges for delegated OAuth * docs(mcp): clarify delegated upstream auth comments * perf(benchmarks): add CPU timing metrics to streaming benchmark (#29980) * Add CPU timing metrics to streaming benchmark * Fix spacing around timing sample dataclass * fix(gemini): don't emit empty choices on metadata-only stream chunks (#29167) web_search + reasoning makes Gemini stream mid-chunks that carry only grounding/thought metadata — no content part, no finishReason. _process_candidates skips content-less candidates and the existing fallback only ran when finishReason was set, so choices stayed empty and the downstream streaming handler raised IndexError on choices[0]. Emit an empty-delta choice for content-less chunks regardless of finishReason. Fixes #28884 * fix(key): allow /key/update to clear budget_limits with [] or null (#30085) * Fix /key/update rejecting budget_limits clear requests with HTTP 400 Sending budget_limits: [] or null to /key/update returned HTTP 400, so once a key had budget windows the last one could never be removed. prepare_key_update_data only json.dumps'd budget_limits when the value was truthy, so [] and None passed through raw to the Prisma Json? column; jsonify_object only serializes dicts, and prisma-client-py has no DbNull sentinel for Json? writes, so Prisma rejected both shapes. Serialize the clear case explicitly as the JSON literal null, matching how memory_endpoints encodes metadata for the same column type. Truthy values keep the existing reset_at window initialization path. Fixes #30067. * Require admin access for budget_limits changes on /key/update Clearing budget_limits via [] or null is a budget mutation, but _validate_update_key_data only counted max_budget and spend as budget changes before deciding whether to skip _check_key_admin_access. A non-admin key owner or a team member with /key/update could therefore remove a key's per-window spend caps without admin authorization. Treat any explicit budget_limits value in the request (set, change, or clear) as a budget change so it gates through the same admin check as max_budget. model_fields_set is used because an explicit null is indistinguishable from an omitted field by value alone. * fix(proxy): persist guardrail info in spend logs for /v1/responses (#30092) Pre-call guardrail blocks on /v1/responses wrote guardrail_information as null in LiteLLM_SpendLogs because _handle_logging_proxy_only_error splits request_data by LoggedLiteLLMParams keys and litellm_metadata, where the Responses API stores request metadata including standard_logging_guardrail_information, was not among them. It fell into optional_params, so merge_litellm_metadata never saw it. Add litellm_metadata to LoggedLiteLLMParams so it routes into litellm_params the same way metadata does on the chat completions path Fixes #28971. * fix(proxy): handle non-standard SSE frames in Anthropic passthrough logging (#26000) Some third-party Anthropic-compatible providers emit non-standard SSE frames (OpenAI-style [DONE] sentinels, non-JSON keep-alive lines) in streaming responses. These caused json.JSONDecodeError in _build_complete_streaming_response, breaking the passthrough logging pipeline so the request was never logged or billed. Skip whole-line 'data: [DONE]' sentinels and catch JSONDecodeError per event. Matching the full line (not a substring) keeps a valid chunk whose text payload contains '[DONE]' from being dropped. Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai> Co-authored-by: Sameer Kankute <sameer@berri.ai> * feat(newrelic): Add New Relic extension (#26989) * initial New Relic integration. * Minor fixes for basic observability. * Implemented basic support for the success path. Generates New Relic custom events needed by the AI Monitorin interface. * Supportability metric is sent on first request. * Emit supportability metric every hour instead of once a day. * Add the start/end times to the messages before sending them so that the start time and end time reflect the correct time and both are not set to 'now'. * Make use of `turn_off_message_logging` configuration that is available by default from CustomLogger. * Enabling New Relic agent to be wired when docker container starts if an environment variable is set. * If we cannot find trace information, send the AI events without the trace ID attached. * Use a fake trace_id if we cannot find one. * Implementing a configuration so that users can use litellm configuration to disable sending LLM messages to New Relic. There is a second method to do this via New Relic env var. * Mised file. * Cleaning up logic to turn off recording content via either the LiteLLM configuration or an env var. * Removing debugging. Fixed logic / comments around how often to send supportability metric. * Initial version of public doc for New Relic. * Use a proper name for the doc file. * Updating newrelic.md document. * Updating LiteLLM documentation for New Relic extension. * Moving New Relic imports into the methods to support unit tests. * Adding unit tests for the New Relic extension. * Updating linting and the unit tests that are not running in the CI environment. * Address reviewer feedback on New Relic integration. - Fix _record_error_metric to use app.record_custom_metric() instead of module-level newrelic.agent.record_custom_metric() so the call works outside of an active transaction context - Remove unreachable except ImportError block in _get_trace_context - Update stale "23 hours" comment to "27 hours" (matches 97200s threshold) - Remove commented-out debug code from _process_success - Fix docs typo: NEW_RELIC_CUSTOM_INSIGHTS_EVENTS_MAX_SAMPLES_STOREDA -> NEW_RELIC_CUSTOM_INSIGHTS_EVENTS_MAX_SAMPLES_STORED - Update TestRecordErrorMetric to verify app.record_custom_metric call Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * Reformating for the linter. * Addressing additional automated feedback. - Removed a legacy comment about the New Relic header - Reordered imports in one file - Switched another file to use the import at the top of the file instead of inline when used - Added unit tests for untested methods that were identified * Addressing new feedback. - Proper handling of time to floats. Created a util method and updated code to use it. - added the missing guard to ensure the app is enabled * Addressing feedback. - When an error occurs, still check if the periodic supportability metric should be emitted - Added a check to ensure the extension is ready in the error handler to match _process_success * Updating the NR event timestamps to more accurately reflect when the messages were generated. * Addressing feedback for potential better practice. * Addressing feedback on accessing default values. Added tests for most of these cases. * Adding a new catch exception block based on feedback. * Addressing feedback about a potential issue around a timestamp for the supportability metric. * Addressing minor feedback on length of generated, fallback traceId. * Addressing feedback. - A few more cases were found where the dictionary access might not return the correct value. - Handling cases where `traceparent` is not lower cased * Addressed feedback where the newrelic options might not apply correctly. * Addressing some feedback. * Addressing feedback. * Validating testing / formatting for our changes. * Updating linting, adding tests, defining data type for UI. * Configuration for the logging callback definition. * Adding a newrelic image for the UI to use. * Putting the New Relic callback in proper alphabetic order. * Copying the logo to a committed output directory so it shows up in a locally built container. * Adding missing definition of new env vars that were causing a build failure. * Addressing automated feedback from greptile. * Adding a few more unit tests to increase the code coverage just a bit more. * Additional unit tests to push coverage to almost 90%. * Adding a custom newrelic docker image build process. This removes the need to add the newrelic agent to the core litellm container or dependencies. * Clarifying message when the New Relic agent is not installed and someone is trying to use the newrelic extension. Either use the proper image when using docker, or install the agent manually when running from source. * Ensuring pip is available to install the New Relic agent. * Updating the definition and handling of traceId (no spanId). Clarifying behavior of env vars vs UI configuration for the newrelic extension. * Removing entries from the New Relic logger configuraiton UI as these values must be set as part of running the image. * Removing a stale doc file that has moved to the litellm-docs repo. Cleanup of Dockerfile to remove a LABEL that was incorrect. * Updating container image name to be the best guess for the new name. * Addressing feedback from greptile. - Added a comment around token_count=0 - Updated the boolean parser to allow a wider set of options which matches existing patterns in other parts of LiteLLM. * Removing option for a separate New Relic container image. The agreement is to handle this in the New Relic integration docs. * Updating error message when New Relic agent is not available. * Wiring in the test message from the LiteLLM callback UX. * Missed saving one of the file conflicts. * Fixed a lint error I introduced. Somehow, I dropped another string and now added it back. * Adding newrelic to the schema definition. * Added an admin check on the call before sending test message as mentioned by the AI code review. * Updating to use should_redact_message_logging(kwargs) as part of the logic to determine if message content should be sent to New Relic or not. This still uses the `record_content` property as well, but both have to be true in order for content to be included. --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com> * Add Azure AI Foundry DeepSeek V3.1 and V4 Pro/Flash global pricing to cost map (#30134) Co-authored-by: Cursor <cursoragent@cursor.com> * fix(logging): translate Responses bridge result to ModelResponse for spend logs (#28985) PR #29394 fixed the AnthropicResponse.model_validate crash for the streaming anthropic_messages -> OpenAI Responses bridge by unwrapping terminal events and returning the inner ResponsesAPIResponse. The spend_logs row lands and usage/cost are correct, but the row's response field stores the Responses API shape (output[...].content[...].text). The proxy UI Logs tab reads response.choices[0].message via parseMessages in prettyMessagesUtils.ts with no fallback for the Responses shape, so the OutputCard renders "No response data available" for every cross-routed call. The same shape mismatch affects every downstream consumer of spend_logs that assumes the canonical chat-completion shape This change keeps the unwrap from #29394 but routes the resulting ResponsesAPIResponse (and the bare-response non-streaming path) through LiteLLMResponsesTransformationHandler.transform_response, which is the same conversion already used by the chat-completion Responses bridge. Spend_logs now stores a ModelResponse with choices[0].message.content, so the UI and other consumers see the assistant text. On a translation failure (eg. empty output on an incomplete response) the handler falls back to a minimal ModelResponse carrying model and usage so the row still lands rather than being dropped as a Non-Blocking error Also corrects a stale comment in the Responses adapter that implied the call type was reclassified to acompletion; the code preserves anthropic_messages and the success handler translates back to ModelResponse for the row Fixes #28595 * fix(anthropic-adapter): re-emit first delta on streaming content-block transitions (#30024) * fix(anthropic-adapter): re-emit first delta on streaming content-block transitions The `/v1/messages` -> `/v1/chat/completions` streaming adapter (`AnthropicStreamWrapper`) silently dropped the first non-empty delta of every content block that started via a *transition* (e.g. text -> tool_use -> text, text -> thinking). When an upstream chunk both triggers a new content block (its type differs from the active block) and carries that block's first delta, the wrapper emitted `content_block_stop` -> `content_block_start` and then only re-queued the trigger chunk when it was an `input_json_delta` (bundled tool args). The synthesized `content_block_start` always carries an empty body, so the first `text_delta` / `thinking_delta` was lost — the client output started from the second token (e.g. "Hi, how can I help you?" rendered as ", how can I help you?", or text resuming after a tool call lost its first sentence). This is especially visible with Claude Code-style clients that consume Anthropic Messages streaming events strictly. Fix: re-queue the trigger chunk's translated delta whenever it carries non-empty content (text/thinking/signature/tool args), via a shared `_trigger_delta_has_content` helper used by both the sync and async paths. Empty trigger deltas are still suppressed so no spurious empty `content_block_delta` is introduced. Fixes #30014 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * test(anthropic-adapter): cover all _trigger_delta_has_content branches Add a direct parametrized unit test for the re-emit predicate so every delta type (text/input_json/thinking/signature), the empty-payload guards, and the malformed/non-delta cases are exercised independently of upstream chunk translation. Raises patch coverage for the new helper. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> * feat: add opt-in healthy_only filter to GET /v1/models (#30130) * feat: add opt-in healthy_only filter to GET /v1/models Adds an opt-in `healthy_only=true` query parameter to GET /v1/models and GET /models that hides models whose backing deployments are all marked unhealthy by background health checks. - Add Router.async_get_fully_unhealthy_model_names(), mirroring the semantics of get_fully_blocked_model_names(): a model is hidden only when every backing deployment is unhealthy and the health state is not stale (fail open otherwise). - Reuses the existing DeploymentHealthCache populated by _run_background_health_check(), so no new health state is introduced. - No-op when allowed_fails_policy is set, mirroring _async_filter_health_check_unhealthy_deployments semantics. - team_public_model_name aliases are aggregated alongside model_name. - Hiding is presentation-only; default behavior is unchanged. Fixes #30128 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs: address Greptile review notes - Note team-alias asymmetry vs get_fully_blocked_model_names - Debug-log when healthy_only is set but no health state is available Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai> Co-authored-by: Claude Fable 5 <noreply@anthropic.com> * Dedupe team soft budget alerts by team_id instead of token (#30097) _team_soft_budget_check sends type="soft_budget" alerts with event_group=TEAM, but SoftBudgetAlert.get_id always returned the request token. The alert cache key was therefore scoped per virtual key, so every active key in a team over its soft budget fired its own alert within budget_alert_ttl. Branch on event_group so team-level alerts dedupe by team_id, matching TeamBudgetAlert, while key and project level alerts keep per-token dedupe. Fixes #27398. * feat(bedrock guardrails): support contextual grounding qualifiers (request-side) (#30057) * test: add failing tests for Bedrock contextual grounding (request-side) Drive the request-side of Bedrock contextual grounding: callers tag message content blocks as grounding_source/query, the post_call hook assembles an ApplyGuardrail(OUTPUT) call carrying source + query + response(guard_content), and the bedrock converse transform must render the tags as prompt text instead of silently dropping them. Non-grounding payloads must stay byte-identical. * feat(bedrock guardrails): support contextual grounding qualifiers Bedrock contextual grounding scores a model response against a reference source and the user query, expressed via a per-content-block `qualifiers` array on ApplyGuardrail. The guardrail hook previously sent plain text only, so grounding could not be driven through it even though the response-side contextualGroundingPolicy parsing already existed. Callers now tag message content blocks `{"type":"grounding_source"}` / `{"type":"query"}` (mirroring the existing `guarded_text` marker). On the generate path the bedrock converse transform renders them as plain text; at post_call the hook harvests them from the request and assembles one ApplyGuardrail(OUTPUT) call carrying grounding_source + query + the response (as guard_content). Requests without these tags produce a byte-identical payload, so existing behaviour is unchanged. * Feat(guardrail): Adding support for custom Ovalix guardrail (#21887) * Feat(guardrail): Adding support for custom Ovalix guardrail * Internal CR comments fixes * greptileai comments fixes * fix conflict * fixes * fix sha256 * clarify Ovalix actor-id hash is for normalization, not PII protection * fix(github_copilot): normalize per-event item_id in /responses streaming (#30072) GitHub Copilot's native /v1/responses stream assigns a different item_id to every event of a single output item (output_item.added, the part.added / delta / done events, and output_item.done). Spec-strict clients like the Vercel AI SDK key streaming parts by item_id and abort with "reasoning part <id> not found" / "text part <id> not found" when a delta references an unregistered id. Override transform_streaming_response in GithubCopilotResponsesAPIConfig to anchor every event of an output item to the id from its output_item.added. Copilot accepts that id paired with the final encrypted_content on the next turn, so multi-turn replay is unaffected. Fixes #30071 * feat: add /model/block and /model/unblock endpoints (#30125) * feat: add /model/block and /model/unblock endpoints Add dedicated proxy-admin POST /model/block and /model/unblock endpoints over the existing blocked flag on LiteLLM_ProxyModelTable, mirroring the /key/block and /key/unblock pattern. Calling a model whose deployments are all blocked now returns a clear 403 "Model is blocked" instead of a generic no-deployment error, including direct-dispatch route types (e.g. eval) via a pre-route guard. Includes audit-log entries for block/unblock and unit tests. Closes #29742 Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> * chore: regenerate dashboard API types for model block/unblock endpoints Regenerate ui/litellm-dashboard/src/lib/http/schema.d.ts from the proxy OpenAPI spec (npm run gen:api) so it includes the new endpoints. Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> * fix: widen router block-helper param type and add direct unit tests Type the _are_all_deployments_blocked deployments parameter to match its callers (DeploymentTypedDict) so mypy passes, and add tests/test_litellm/test_router_block_helpers.py with direct unit tests for the three block helper methods so router_code_coverage recognizes them. Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> * fix: restore type-ignore on messages arg after black reflow Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> * refactor: raise model-block 403 in proxy layer, not SDK Router Keep the SDK Router's documented behavior for blocked deployments (filtered -> "no healthy deployment") and move the 403 PermissionDeniedError into the proxy layer (route_llm_request), where model blocking is an admin concept. This avoids a backwards-incompatible 403 for SDK users who set blocked=True on their own deployments, per maintainer review. Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> --------- Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> Co-authored-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> Co-authored-by: Sameer Kankute <sameer@berri.ai> * fix: add week unit support to get_next_standardized_reset_time (#30100) * fix: add week unit support to get_next_standardized_reset_time The function handled d/h/m/s/mo units but silently fell through to the default next-midnight branch for the w (week) unit. This was inconsistent: _extract_from_regex already accepted w in its character class, and duration_in_seconds already returned value * 604800 for it. Add the missing elif unit == 'w' branch that delegates to _handle_day_reset with value * 7, which reuses the existing Monday- alignment logic for 1w and the generic N-day-from-midnight path for larger multiples. Add test_week_based_resets covering 1w from a Wednesday (expects next Monday) and 2w from a Monday (expects 14 days forward at midnight). Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> * test: exercise relative week semantics with non-Monday base dates + add docstring Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> --------- Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> Co-authored-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> * fix: black formatting and remove undocumented MAVVRIK_FOCUS_FREQUENCY env var * fix: black formatting with correct version and sync schema.d.ts for healthy_only param * fix: resolve mypy errors and add transcription_sessions to JSON schema endpoint enum * fix: restore MAVVRIK_FOCUS_FREQUENCY guard and exclude it from docs key scan * fix: address Greptile P2 comments - move constant, use UTC datetime, skip redundant team lookup * revert: restore original team lookup logic in can_key_call_resolved_model --------- Signed-off-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> Signed-off-by: FugoP <264910004+AgentGymLeader@users.noreply.github.com> Co-authored-by: Emerson Gomes <emerson.gomes@thalesgroup.com> Co-authored-by: nina-hu <nina.huuu@gmail.com> Co-authored-by: Sahith Jagarlamudi <104647530+s-jag@users.noreply.github.com> Co-authored-by: shin-berri <shin-laptop@berri.ai> Co-authored-by: yuneng-jiang <yuneng@berri.ai> Co-authored-by: Praveen Ghuge <95286176+pghuge-cloudwiz@users.noreply.github.com> Co-authored-by: alex107ivanov <30668368+alex107ivanov@users.noreply.github.com> Co-authored-by: hcl <chenglunhu@gmail.com> Co-authored-by: Fede Kamelhar <federico.kamelhar@oracle.com> Co-authored-by: Armaan Sandhu <74664101+Ar-maan05@users.noreply.github.com> Co-authored-by: Teo Xian Zhong Augustine <35527068+auggie246@users.noreply.github.com> Co-authored-by: King Star <mcxin.y@gmail.com> Co-authored-by: Saksham Maggo <122939011+SakshamMaggo@users.noreply.github.com> Co-authored-by: Filippo Menghi <113345637+Cyberfilo@users.noreply.github.com> Co-authored-by: Kelvin <leikaiwei@outlook.com> Co-authored-by: Josh Bonczkowski <josh.bonczkowski@gmail.com> Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com> Co-authored-by: M. Dennis Turp <mdturp@pm.me> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Piotr Minkina <piotrminkina@users.noreply.github.com> Co-authored-by: Martín Alcalá Rubí <martin@tryolabs.com> Co-authored-by: T. Kobayashi <13004314+nix-tkobayashi@users.noreply.github.com> Co-authored-by: João Costa <13508071+jpv-costa@users.noreply.github.com> Co-authored-by: Shalom <shalom@ovalix.io> Co-authored-by: codgician <15964984+codgician@users.noreply.github.com> Co-authored-by: FugoP <kim@pomsora.com> Co-authored-by: AgentGymLeader <264910004+AgentGymLeader@users.noreply.github.com> Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>	2026-06-11 22:30:26 -07:00
Yassin Kortam	012d9f6c0a	feat(rate-limiter): allow opting out of v3 TPM reservation and Redis circuit breaker (#30211)	2026-06-11 10:34:26 -07:00
ryan-crabbe-berri	0d120de785	chore(hooks): enforce Conventional Commits and Conventional Branches (#30174) ... * chore(hooks): enforce Conventional Commits and Conventional Branches Adds opt-in local git hooks plus a CI PR-title check: - .githooks/commit-msg validates commit subjects against Conventional Commits 1.0.0 (feat|fix|docs|style|refactor|perf|test|build|ci| chore|revert)(scope)!: subject. Merge/revert/fixup!/squash!/amend! messages pass through; --no-verify still works. - .githooks/pre-push validates branch names against Conventional Branches (feature|bugfix|hotfix|release|chore)/desc. Bypasses main, litellm_internal_staging, dependabot/*, gh-readonly-queue/*. Tag pushes and deletions are skipped. - scripts/install_git_hooks.sh sets core.hooksPath=.githooks and is wired up as 'make install-hooks'. Opt-in — not chained into install-dev. - .github/workflows/conventional-commits.yml validates PR titles via amannn/action-semantic-pull-request pinned to v6.1.1's SHA. This is the actual gate since squash-merge uses the PR title as the commit subject. - tests/test_litellm/test_git_hooks.py exercises both hooks via subprocess for accept / reject / bypass / git-generated-message cases. - CONTRIBUTING.md documents the conventions, the install step, the bypass list, and the --no-verify escape hatch. Resolves LIT-3306 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(hooks): address Greptile review on PR #28703 Resolves two findings from the automated code review: 1. CONTRIBUTING.md: shrink the new Conventional Commits / Branches section to a 2-line pointer at docs.litellm.ai. Per the team convention, the full documentation lives in the litellm-docs repo — see BerriAI/litellm-docs#208 for the companion change that adds the section to docs/extras/contributing_code.md. 2. .githooks/commit-msg: tighten the subject regex to also reject an uppercase first letter in the description. CI's subjectPattern is ^(?![A-Z]).+$ so the previous local hook would accept 'feat: Add thing' which would then fail the PR-title check. The local hook is now the strictly tighter of the two gates. Test cases extended to cover both the new rejection and the digit/symbol-start cases that remain allowed. Resolves LIT-3306 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * chore: trigger ci after branch rename * fix(ci): rerun pr title check when bypass label changes amannn/action-semantic-pull-request only honors ignoreLabels if the workflow retriggers on labeled/unlabeled events; without them a red check stays red after a maintainer applies the bypass label. Also point the CONTRIBUTING.md workflow comments at the conventions section, which now sits above the Development Workflow section. --------- Co-authored-by: Yassin Kortam <yassinkortam@Yassins-MBP.localdomain> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-06-11 10:00:23 -07:00
Mateo Wang	20e453f698	feat(cli): per-agent lite claude / codex / opencode commands that wrap coding agents through the proxy (#29850) ... * feat(cli): add `litellm-proxy run -- <agent>` to wrap coding agents through the proxy Wraps Claude Code, Codex, OpenCode, and any other coding agent so all of its LLM traffic routes through a LiteLLM proxy, with the agent-vault style of "just works" DX: one `run -- <agent>` command, auto SSO login when interactive, env-key "agent mode" for containers/CI, and a fail-fast key check against the proxy so bad credentials error immediately instead of deep inside the agent. The wrapped binary is detected by name to pick the right variables. Claude Code gets ANTHROPIC_BASE_URL (the bare proxy root, so it appends /v1/messages) and ANTHROPIC_AUTH_TOKEN, with any stray ANTHROPIC_API_KEY cleared so the proxy token wins. Codex and OpenCode get OPENAI_BASE_URL (proxy + /v1) and OPENAI_API_KEY. Unrecognized commands get both sets so they work either way. `litellm-proxy claude-code` remains as a shortcut for `run -- claude`. The core logic is split into dependency-injected helpers (agent_profile, build_agent_env, verify_proxy_key, run_agent) so env wiring, the preflight, and the launch handoff are unit-tested without monkeypatching, alongside CliRunner tests for auth resolution, agent mode, and auto-login. Mutation-tested the env profiles, preflight, and agent-mode branch to confirm the tests fail when the behavior is broken. https://claude.ai/code/session_0154VpLXW7mMvk5wfbgPRJa6 * Make each coding agent its own litellm-proxy command Replace the `run -- <agent>` interface and the `claude-code` shortcut with top-level commands generated per known agent, so launching is just `litellm-proxy claude`, `litellm-proxy codex`, or `litellm-proxy opencode`, with everything after the agent name forwarded straight to it. This drops the ceremony of `run --` and cuts typing. The `--model`/`--small-fast-model` wrapper flags are gone; pass the agent's own model flag instead, or export the model env vars (the wrapper preserves what you already have set), which keeps the surface minimal and avoids intercepting flags the agent owns. Rename the module to agents.py to match. * fix(cli): route `litellm-proxy codex` through the proxy via a custom provider Codex ignores OPENAI_BASE_URL (it always dials api.openai.com over the Responses WebSocket transport), so the OpenAI env profile alone left `litellm-proxy codex` talking to OpenAI directly instead of the proxy. Point Codex at the proxy with a custom provider passed as `-c` config overrides, and force the HTTP/SSE Responses transport with supports_websockets=false since the proxy does not speak the Responses WebSocket protocol. The provider reads its key from OPENAI_API_KEY, which the agent env already exports. The overrides are injected ahead of the user's args so they precede Codex's subcommand. Claude Code and OpenCode are unaffected; they honor the exported env vars. Adds regression tests for the per-agent launch args and the injection ordering. Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com> * Rename litellm-proxy CLI command to lite The proxy management CLI was invoked as litellm-proxy, which is a lot to type for an everyday command. Rename the console script entry point to lite and update the in-CLI usage examples, help text, error messages and docs to match. * fix(sso): stop CLI auth success page from hanging on "Closing..." The CLI opens the SSO success page with webbrowser.open, so the tab is not script-opened and the browser refuses window.close(). The countdown would end on "Closing..." and the tab would sit there forever. Drop the countdown and just show "You can now close this window and return to your terminal." from the start, while still attempting window.close() once so the tab auto-closes in the rare case the browser allows it. Add a regression test asserting the manual-close instruction is always present and the misleading countdown/"Closing..." text is gone. * fix(cli): reattach controlling terminal after SSO login, keep litellm-proxy alias When the first `lite claude` has to log in via browser SSO, completing the login could leave stdin detached from the terminal, so a TUI agent like Claude Code would start in non-interactive mode and exit with "Input must be provided". The wrapper now reopens the controlling terminal onto stdin just before handoff when the session started interactively; piped or redirected input is detected up front and left alone, so agent-mode and non-interactive use are unchanged. Also keep the `litellm-proxy` console script as an alias for `lite` so existing scripts and CI that invoke `litellm-proxy` keep working; both names map to the same CLI. * feat(install): make the curl installer need only curl, not a pre-existing Python The installer now lets uv provision a managed Python 3.13 when no suitable interpreter is found, instead of aborting. The minimum is also bumped from 3.9 to 3.10 to match the package's requires-python (>=3.10), so a system Python 3.9 is no longer selected only for uv tool install to reject it. * feat(cli): add thin litellm[cli] install path (install-cli.sh + brew) for the lite CLI On a developer laptop the `lite` CLI only needs `lite login` and running coding agents through a proxy, but the sole install path was `litellm[proxy]`, which drags in the whole server tree (fastapi, uvicorn, boto3, polars, cryptography, litellm-enterprise). The CLI's heavy imports are all guarded, so it runs on the base SDK plus just rich, pyyaml and requests. Add a `cli` extra carrying exactly those three, a `scripts/install-cli.sh` curl one-liner that installs `litellm[cli]`, and a `BerriAI/homebrew-litellm` tap formula with a release runbook under `packaging/homebrew/`. The installer passes no `--python`, so uv honours litellm's requires-python and provisions a managed interpreter, skipping a too-old (3.9) or too-new (3.14+) system Python instead of failing to resolve. A pyproject thin-contract test asserts the `cli` extra keeps the deps the CLI imports and never leaks a server-only dependency from `proxy`, so the laptop install cannot silently re-bloat * fix(install): let uv pick the Python via --python-preference system Both installers detected a system Python with a floor-only check and forced it with `uv tool install --python <interp>`. On a host whose only Python is outside litellm's requires-python (a too-old 3.9 or, increasingly, a too-new 3.14) that forced an incompatible interpreter and the resolve failed. Drop the detection and pass `--python-preference system`: uv reuses a compatible system Python when present and downloads a managed one otherwise, always honouring requires-python * test(router): filter aiohttp unclosed-session gc noise in test_async_fallbacks test_async_fallbacks asserts the last three captured log records are the router's fallback messages. Under the litellm_router_testing job (pytest -k router -n 4) many router tests share the module-level in_memory_llm_clients_cache (max 200, ttl 3600s). Older cached OpenAI/Azure clients get evicted while their aiohttp ClientSession is still open, and when the gc reclaims them aiohttp emits "Unclosed client session"/"Unclosed connector" through the asyncio logger. Those records land in caplog mid-test and push the expected router logs out of the last-three window, so the assertion flips to failing non-deterministically. These warnings are async cleanup noise, not router debug logs, so filter them out exactly like the existing leaked-task warnings before asserting order. The assertion on the three router fallback messages is unchanged. --------- Co-authored-by: Cursor Agent <cursoragent@cursor.com> Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com> Co-authored-by: Claude <noreply@anthropic.com>	2026-06-10 13:52:26 -07:00
Yassin Kortam	d5d6b26a72	fix: improve bedrock streaming hot path perf (#28720)	2026-05-28 11:31:37 -07:00
Yassin Kortam	2eab9ee2c0	perf: reduce per-request and per-chunk overhead across Anthropic streaming hot paths (#28289) ... * perf: reduce per-request and per-chunk overhead across Anthropic streaming hot paths - Introduce pure-text fast-path in `_build_complete_streaming_response` that collapses O(N) `content_block_delta` events into a single equivalent SSE event before conversion, eliminating per-output-token Pydantic `ModelResponseStream` construction; non-text streams (tool_use, thinking, citations) fall back to the unchanged legacy path - Skip agentic streaming wrapper entirely when no callback overrides `async_should_run_agentic_loop`; the wrapper buffered every chunk and rebuilt the SSE response only to call hooks that all return `(False, {})` — a pure no-op for the default config - Serialize request body once (`json.dumps`) for both the pre-call log input and the wire, instead of twice; avoids a full O(payload) scan per request, significant for long-context Claude Code histories - Add fast path in `async_streaming_data_generator` that bypasses the per-chunk `async_post_call_streaming_hook` coroutine await, response-string materialization, and cost-injection call when no callback/guardrail/cost-injection is active (the default config) - Resolve `_DD_STREAMING_TRACE_ENABLED` once at import time; eliminate per-chunk `NullSpan` context manager allocation when Datadog tracing is disabled (the default) - Memoize `get_type_hints(AnthropicMessagesRequestOptionalParams)` with `@lru_cache(maxsize=1)` — resolves once per process instead of once per `/v1/messages` request (~80µs each) - Hoist `cost_injection_active` out of the per-chunk loop in `chunk_processor`; eliminates repeated `getattr` + endpoint-type checks on every streamed byte chunk - Extract `_build_passthrough_logging_result` from `_route_streaming_logging_to_handler` as a standalone static method to facilitate future off-loop dispatch - Convert `async_sse_data_generator` from an `async for: yield` trampoline to a direct return of the underlying generator, removing one async-generator layer per streamed chunk - Skip redundant `strip_empty_text_blocks_from_anthropic_messages` scan in `anthropic_messages_handler` when the async wrapper already sanitized (signalled via `_litellm_messages_presanitized` sentinel, popped before reaching provider params) - Gate debug log `f-string` evaluation behind `isEnabledFor(DEBUG)` in both the streaming generator and the transformation layer to avoid serializing entire message payloads on every request at non-debug log levels - Add benchmark script (`scripts/benchmark_anthropic_messages_perf.py`) with a local mock Anthropic SSE provider for reproducible TTFT and TPM measurement across commits/branches - Add parity tests asserting fast-path and legacy-path produce byte-identical logged/billed payloads, plus unit tests for agentic hook detection, pre-serialized body reuse, and memoized key resolution * perf: address greptile review for anthropic streaming hot path - Bail to legacy in `_collapse_pure_text_chunks` when content_block_delta events from different block indexes are observed without an intervening flush. Anthropic sends blocks strictly sequentially, but defensive bail prevents silent text-merging if the protocol ever interleaves. - Replace leaf-class `__dict__` check for `async_post_call_streaming_hook` in `_callback_capabilities` with a function-identity comparison that walks the MRO. A vendor base class can carry the override and the registered class can add nothing else; before this PR the hook was unconditionally invoked, so an inherited-override miss would silently drop the hook on the streaming path. - Add unit tests for both behaviors. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(mypy): narrow model_name to str in cost-injection branch The hoisted cost_injection_active flag in chunk_processor encodes the `bool(model_name)` requirement but mypy can't track that invariant through the local, so the per-chunk `_process_chunk_with_cost_injection( chunk, model_name)` calls flagged Optional[str] vs str. Pin a typed non-None local inside the cost-injection branch so mypy narrows correctly without changing runtime behavior. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Yassin Kortam <yassinkortam@g.ucla.edu> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-23 12:15:59 -07:00
Yassin Kortam	a6494e6fe3	perf: eliminate per-request callback scanning on proxy hot path (#27858) ... - Introduce `_CallbackCapabilities` dataclass and `ProxyLogging._callback_capabilities()` static method that inspects `litellm.callbacks` once and caches capability flags keyed on (list length, member ids); invalidates automatically when the callback list mutates without per-request iteration overhead - Replace O(n) `litellm.callbacks` walks in `async_pre_call_hook`, `during_call_hook`, `async_post_call_streaming_iterator_hook`, `async_post_call_streaming_hook`, and `post_call_response_headers_hook` with fast-path exits when no relevant callbacks are registered - Add `needs_iterator_wrap()` and `needs_per_chunk_streaming_hook()` instance methods to decouple iterator-level wrapping from per-chunk hook execution; avoids `get_response_string` materialization per chunk when no guardrail or chunk-hook callback is active - Introduce `_fast_serialize_simple_model_response_stream()` using `orjson` for common single-choice text streaming chunks, bypassing the full Pydantic serializer; falls back to `model_dump_json` for tool calls, logprobs, usage, and provider-specific fields - Add early-return in `_restamp_streaming_chunk_model` when downstream model already matches the requested model, avoiding unnecessary string comparisons on every chunk - Fix stale zero-cost cache bug in `_is_model_cost_zero`: move the per-router `_zero_cost_cache` dict onto the `Router` instance and clear it in `_invalidate_model_group_info_cache` so in-place pricing updates via `upsert_deployment` immediately resume budget enforcement - Add `scripts/benchmark_chat_completions_perf.py`: standalone async benchmarking tool with a mock OpenAI provider, LiteLLM proxy process management, non-streaming RPS, streaming TTFT, and full-stream latency measurements with repeat/median run support - Add comprehensive unit tests covering capability detection, cache invalidation, fast-path correctness, zero-cost cache regression, and the no-callback streaming fast path Co-authored-by: Yassin Kortam <yassinkortam@g.ucla.edu>	2026-05-14 09:28:31 -07:00
ryan-crabbe-berri	be84d5cd7d	ci: add manually-triggered mutation testing workflow (#27576) ... * ci: add manually-triggered mutation testing smoke workflow Adds a workflow_dispatch-only GitHub Actions workflow that runs mutmut against a single source/test pair (router_settings_endpoints) to validate the tooling end-to-end before scaling. The workflow reinstalls litellm non-editable so the mutants/ sandbox is not shadowed by the editable .pth on sys.path, and sets PYTHONPATH so the trampolined sandbox copy wins over site-packages. mutmut itself is pulled in via uv run --with so it does not appear in uv.lock or affect the shared dev environment. Includes a temporary push: trigger scoped to this branch so we can iterate before the workflow file lands on the default branch — to be removed before merging (workflow_dispatch only requires the file on the default branch to surface the manual trigger button). * ci(mutation): disable rerun and xdist plugins for mutmut runs mutmut's in-process pytest.main() call hits `INTERNALERROR: no option named 'filtered_exceptions'` from pytest-retry's pytest_configure hook. Reruns are also wrong for mutation testing — a "failed" mutant test that gets retried would mask which mutants are killed vs. survive. Disable retry, rerunfailures, and xdist via pytest_add_cli_args in [tool.mutmut]. * ci(mutation): uninstall pytest-retry before mutmut runs `-p no:retry` (and similar names) didn't match pytest-retry's entry-point name, so the plugin still loaded and crashed during mutmut's "Running clean tests" phase. Uninstalling the package is surgical and doesn't depend on guessing the entry-point name. * ci(mutation): emit per-survivor diffs to run-page summary + artifact The previous artifact only contained `mutmut results` text (which in mutmut 3.x lists survivor names but not the actual mutations). Adds: - `mutmut export-cicd-stats` to produce mutmut-cicd-stats.json with the killed/survived/total scoreboard. - `mutmut show <name>` per surviving mutant to capture each mutation as a unified diff. - A `mutmut-report.md` that combines summary + run-progress tail + per-survivor diffs, written to both the artifact and $GITHUB_STEP_SUMMARY (visible on the run page, no download needed). - Corrected artifact paths: stats files live under mutants/, not the project root. - The trampolined source file from the sandbox so survivors can be inspected even outside `mutmut show`. * ci(mutation): document intended manual weekly cadence in trigger comment * ci(mutation): generate ACH-style report with embedded function bodies Replaces the inline bash markdown generation with a Python script that: - Groups survivors by function (one section per function, function body shown once per section, surviving mutants nested as subsections) - Embeds each enclosing function's source via Python AST (so the agent has full context, not just a 3-line `mutmut show` diff) - Inlines the existing test file(s) listed in [tool.mutmut].tests_dir - Writes an ACH-style task description at the bottom following the prompt template from arXiv 2501.12862 Output goes to mutation-report.md (artifact) and the head of the file is appended to $GITHUB_STEP_SUMMARY for at-a-glance visibility. * fix(mutation report): correctly parse function names with leading underscores mutmut's mutant-name prefix is x_ (single underscore), so a function named _foo produces mutants x__foo__mutmut_N. The previous regex \.x__(.+)__mutmut_ ate the function's leading underscore as part of the prefix. Changed to \.x_(.+)__mutmut_ so leading underscores are preserved in the captured function name; verified for normal, leading- underscore, and dunder-method names. * feat(mutation report): full Meta ACH-style rendering with MUTANT delimiters For each surviving mutant, parse the mutmut sandbox trampoline file and render the mutated function as it appears in the source — with the differing lines wrapped in `# MUTANT START` / `# MUTANT END` comments, matching the format from Meta's ACH paper (arXiv 2501.12862, Table 1). Renames the function header back to its original name so the agent sees the function as it would appear in the file. Falls back to the unified diff if the trampoline lookup fails. Handles replace, insert, and delete diff ops; uses difflib's SequenceMatcher to find the differing line ranges. The unified diff is preserved in a collapsible <details> block as secondary context. * ci(mutation): scope to whole management_endpoints folder, drop temp push trigger Final scope before merge: - paths_to_mutate / tests_dir broadened from one file to the entire management_endpoints source/test folders - Trigger is now `workflow_dispatch` only — the temporary push: block used during workflow iteration is removed - timeout-minutes bumped from 60 to 350 (just under the GH-hosted job cap of 360); whole-folder mutation against ~15 files / ~7.5k LOC can take a few hours - Artifact path for the trampoline files glob-expanded to cover all files under mutants/litellm/proxy/management_endpoints/ * fix(mutation report): warn when multiple functions in a file share a name Addresses the Greptile review concern: ast.walk's first-match-wins behavior could embed the wrong function body when a file defines the same name in multiple places (e.g., a module-level helper and a class method). mutmut's mutant identifier does not carry class context, so we can't always determine which definition was mutated. find_function_in_file now returns the start line of every matching definition; render() surfaces a "Note: N functions named X" warning in the report when there is more than one match. The first match is still embedded as the body — the warning tells the reader to verify manually instead of silently using the wrong context. Smoke-tested against the existing artifact: single-match files render unchanged. * Fix mutation report anchors * Fix mutation report TOC anchors --------- Co-authored-by: Cursor Agent <cursoragent@cursor.com>	2026-05-11 15:19:57 -07:00
harish-berri	a67b7a7e87	Refactor Bedrock response stream shape handling (#27257) ... * Refactor Bedrock response stream shape handling - Introduced a module-level constant `BEDROCK_RESPONSE_STREAM_SHAPE` to cache the response stream shape, eliminating the need for per-instance caching in `BedrockEventStreamDecoderBase`. - Updated relevant methods to utilize the new constant, improving performance by avoiding redundant loading of the shape. - Added tests to ensure the shape is loaded correctly at import time and is consistent across different modules. - Added a new mock server script for testing Bedrock pass-through functionality. * Refactor response parsing for Bedrock and SageMaker - Improved code readability by formatting the parsing method calls in `AWSEventStreamDecoder` for both Bedrock and SageMaker response stream shapes. - Added blank lines for better separation of code blocks in `invoke_handler.py` and `common_utils.py` to enhance maintainability. * Enhance error handling for Bedrock and SageMaker response stream shape loading - Wrapped the loading logic in `_load_bedrock_response_stream_shape` and `_load_sagemaker_response_stream_shape` with try-except blocks to gracefully handle exceptions. - Added logging to warn when the response stream shape cannot be pre-loaded, ensuring the module imports cleanly. - Updated tests to verify that loading failures return `None` instead of propagating exceptions. * Implement error handling for missing response stream shapes in Bedrock and SageMaker - Added checks in `_parse_message_from_event` methods to raise appropriate errors when `BEDROCK_RESPONSE_STREAM_SHAPE` or `SAGEMAKER_RESPONSE_STREAM_SHAPE` is None, ensuring clearer error reporting. - Updated logging messages to reflect the unavailability of event-stream decoding for both Bedrock and SageMaker. - Enhanced unit tests to verify that the correct exceptions are raised when the response stream shapes are not loaded.	2026-05-06 17:39:38 -07:00
Yassin Kortam	950074eea2	fix: atomic TPM rate limit (#27001) ... Co-authored-by: Yassin Kortam <yassinkortam@g.ucla.edu>	2026-05-05 16:58:07 -07:00
Krrish Dholakia	b6fc75b3ce	Merge branch 'litellm_internal_staging' into litellm_adaptive_routing	2026-04-20 15:28:08 -07:00
Krrish Dholakia	386f334fee	Prompt Compression - add it to the proxy (#25729) ... * refactor: new agentic loop event hook simplifies how to create logic for tool based multi llm calls * fix: compress - make it work on anthropic input as well * fix(compress.py): working prompt compression for claude code ensures claude code messages can run through proxy easily * docs: add agentic loop hook guide * docs: add agentic_loop_hook to sidebar * fix: fix multiple arguments error * fix: fix tool call loop for compression on streaming /v1/messages * fix: fix linting errors * fix: fix ci/cd errors * feat(litellm_pre_call_utils.py): use claude code session for litellm session id allows claude code logs to be stitched together, making it easy to know they were all part of the same conversation * fix: suppress incorrect mypy warning rE: module * revert: drop PR's changes to litellm/proxy/_experimental/out/ Restores the 34 HTML files under _experimental/out/ to their pre-PR paths (X/index.html -> X.html). All renames are R100 (content unchanged); no other files are touched. * fix: address greptile review comments on PR #25729 - Skip ``kwargs["tools"] = []`` injection when compression is a no-op — Anthropic Messages rejects empty tool arrays on requests that did not originally declare tools. - Move agentic-loop safety guards (fingerprint cycle / max depth) out of the per-callback try/except so they propagate instead of being swallowed by the generic exception handler. Extracted _check_agentic_loop_safety. - Gate generic ``x-<vendor>-session-id`` capture behind the LITELLM_CAPTURE_VENDOR_SESSION_HEADERS env var (off by default) to preserve backwards compatibility; explicit x-litellm-* headers are unaffected. - Fix monkeypatch target in pre-call-hook test to patch the actual module-level binding (litellm.integrations.compression_interception.handler.compress). - Add regression tests for empty-tools skip and opt-in session capture. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * revert: drop LITELLM_CAPTURE_VENDOR_SESSION_HEADERS flag Generic x-<vendor>-session-id header capture is a new feature and only runs *after* the explicit x-litellm-trace-id / x-litellm-session-id checks, so it does not change behavior for any existing caller that was already using the LiteLLM headers — no backwards-incompatibility to gate. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * refactor(compress): replace input_type with CallTypes call_type Drop the bespoke ``CompressionInputType`` literal and use the existing ``litellm.types.utils.CallTypes`` enum instead. ``litellm.compress()`` now takes ``call_type: Union[CallTypes, str]`` (default ``CallTypes.completion``) — no new concept to learn, and the enum is already the way the rest of the codebase talks about request shapes. Supported values: ``completion`` / ``acompletion`` (OpenAI chat-completions shape) and ``anthropic_messages`` (Anthropic structured content blocks). Updated: compress(), the compression_interception handler, tests, docs, and the two eval scripts. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-20 15:08:00 -07:00
Krrish Dholakia	924fa6a3bc	feat: commit new adaptive routing	2026-04-18 21:29:39 -07:00
Krrish Dholakia	dd4a1d2be2	feat: add adaptive routing to litellm ... allow model routing to improve based on conversation signals ensures router is picking best model for task	2026-04-18 16:35:17 -07:00
Ishaan Jaffer	e8461b5b97	style: run black formatter on files from main merge	2026-04-17 13:02:59 -07:00
ishaan-berri	cb8fc480e6	Merge pull request #25732 from harish876/health-check-oom ... Optimize database query to prevent OOM errors during health checks	2026-04-15 18:13:11 -07:00
harish876	d20c70f24c	Optimize database query which fetches latest model_id, model_name pairs and dedupes them in memory. ... Current fix includes - Updates test case - Optimized query with docstring. The change leverages deduplication and sorting logic from SQL - Added a bench script to differentiate peak memory usage before and after	2026-04-15 00:54:37 +00:00
ishaan-berri	0e43050a01	Merge pull request #25650 from BerriAI/litellm_dev_04_13_2026_p1 ... feat: add litellm.compress() — BM25-based prompt compression with ret…	2026-04-14 12:24:47 -07:00
Krrish Dholakia	26c7412339	feat: add litellm.compress() — BM25-based prompt compression with retrieval tool (#25637) ... * feat: add litellm.compress() for BM25-based context compression Adds a compress() utility that reduces context size for LLM calls using BM25 relevance scoring (with optional semantic embeddings via litellm.embedding()). Messages below a token threshold pass through unchanged; messages above are scored, ranked, and the lowest-relevance ones replaced with stubs. Originals are cached and a retrieval tool is injected so the model can recover dropped content on demand. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(compress): truncate high-scoring messages instead of fully stubbing them When a relevant message was too large to fit in the token budget it was replaced with a stub, leaving the LLM with no real content to work with. Now the highest-scoring overflow message is truncated (first 70% + last 30% of words) to fill the remaining budget, so the LLM always receives actual content rather than just a retrieval pointer. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(bm25): add prefix expansion so query terms match inflected doc tokens "cook" now matches "cooking", "auth" matches "authentication", etc. Without this, short query terms scored 0 against longer inflected forms in documents, causing the wrong message to be kept. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * test: add routing correctness test and eval harness for litellm.compress() - test_simple_compression: parametrized test verifying BM25 routes the right message based on query ("How to cook?" keeps cooking, "Fix auth" keeps auth content) - eval_compression.py: end-to-end eval harness comparing baseline vs compressed model performance on HumanEval-style coding problems Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat(eval): add SWE-bench Lite compression eval harness Uses princeton-nlp/SWE-bench_Lite_bm25_27K which bundles ~27k tokens of BM25-retrieved repo context per problem — large enough to meaningfully stress litellm.compress() without Docker or GitHub API calls. Proxy eval metrics (no test runner needed): - has_diff: model produced a valid unified diff - file_overlap: fraction of gold-patch files in generated patch - exact_file_match: generated patch touches exactly the right files Run: python tests/eval_swe_bench.py --model gpt-4o --problems 10 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(eval): robust dataset loading + sys.path fix for worktree imports - Add HuggingFace API fallback so the SWE-bench loader doesn't need the `datasets` library (avoids pyarrow/numpy binary compat issues) - Insert repo root into sys.path so compression module resolves from worktrees - Use direct import of litellm_compress to avoid __getattr__ issues Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * improve compression quality: line-based truncation, multi-message budget, 70% default target - Switch truncate_message from word-based to line-based splitting to preserve code structure (function boundaries, indentation) - Allow multiple messages to be truncated instead of burning entire budget on one overflow message - Raise default compression target from 50% to 70% of trigger for better quality/cost tradeoff - Add --compression-target CLI arg to SWE-bench eval harness - Move tests to canonical locations (tests/test_litellm/, scripts/) - Add docs page and sidebar entries for compress() Eval results (5 problems, Opus, trigger=10k): Hunk overlap delta improved from -0.417 to -0.221 Content similarity now matches baseline (+0.006) Cost savings: 72% Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * docs: add SWE-bench performance results to compress() docs Include benchmark table from Opus eval (5 problems, trigger=10k) showing 72% cost savings with file-level quality fully preserved. Add metric explanations and eval runner examples. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(eval): use tolerance-based hunk overlap metric The exact line-number matching was too brittle — LLM-generated patches often target the right code region but with slightly offset line numbers. Switch to hunk-level overlap with a 10-line tolerance window so nearby edits count as matches. This better reflects actual patch quality. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: add compression_interception callback for LiteLLM Proxy Add a proxy callback that automatically compresses incoming /v1/messages payloads above a configurable token threshold, runs the retrieval tool loop server-side, and returns the final response. This brings compress() support to proxy deployments (e.g. Claude Code via /v1/messages). - New callback: litellm/integrations/compression_interception/ - Proxy config: compression_interception_params in litellm_settings - Support for input_type param in compress() (openai vs anthropic) - Docs: proxy setup instructions with YAML config example - Tests: 139-line unit test suite for the interception handler Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * Revert "feat: add compression_interception callback for LiteLLM Proxy" This reverts commit 72bd5cb152ca1df07f14a14e14a2816e188874a8. --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-13 12:23:54 -07:00
stuxf	a6c30b30bf	build: migrate packaging, CI, and Docker from Poetry to uv (#25007) ... * build: migrate packaging metadata to uv * ci: move automation and local tooling to uv * docker: migrate image builds and runtime setup to uv * docs: update install and deployment guidance for uv * chore: align auxiliary scripts and tests with uv * test: harden test_litellm isolation * fix: keep release and health check images self-contained * build: pin uv tooling and health check deps * test: isolate bedrock image request formatting from suite state * test: cover sandbox executor requirements flow * ci: fix circleci no-op command steps * ci: fix circleci publish workflow parsing * fix: stabilize remaining uv migration CI checks * ci: increase matrix test timeout headroom * fix: restore published docker and license coverage * fix: restore proxy runtime build parity * fix: restore proxy extras parity and venv migrations * ci: persist uv path across circleci steps * fix: keep psycopg binary in default test env * docker: preserve prisma cache across stages * test: run local proxy checks through uv python * build: restore runtime deps moved into ci * build: refresh uv lock after upstream merge * fix: restore module import in test_check_migration after merge The conflict resolution imported only the function but the test body references check_migration as a module throughout. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: revert dependency promotions, remove nodejs-wheel-binaries, fix Docker layer caching - Move google-generativeai, Pillow, tenacity back to ci group (they are lazily imported and bloat the base SDK install needlessly) - Remove nodejs-wheel-binaries from extra_proxy and proxy-dev (redundant in Docker where system Node.js is already installed via apk) - Remove all nodejs-wheel node replacement and venv npm patching blocks from Dockerfiles since the wheel is no longer installed - Add --no-default-groups to CodSpeed benchmark workflow so the benchmark environment matches the old minimal pip install footprint - Apply standard uv two-phase Docker pattern: copy metadata first, install deps (cached layer), then copy source and install project - Replace CircleCI enterprise no-op with proper uv sync command Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * chore: regenerate uv.lock after removing nodejs-wheel-binaries Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(ci): use cache/restore instead of cache to prevent cache poisoning The old workflow used actions/cache/restore (read-only). The uv migration changed it to actions/cache (read-write), which zizmor flags as a cache poisoning risk. Restore the safer read-only variant. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(ci): disable setup-uv built-in cache to silence cache-poisoning alert The setup-uv action enables caching by default, which zizmor flags as a cache poisoning risk. Disable it since we already use a read-only cache/restore step. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(ci): disable setup-uv cache in publish workflow Silences zizmor cache-poisoning alert. Publishing workflow runs infrequently on protected branches so caching adds no real benefit. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(test): remove duplicate verbose_logger mock in test_check_migration The logger was patched twice — first via mocker.patch() then via mocker.patch.object(autospec=True). The second call fails because autospec cannot inspect an already-mocked attribute. Remove the redundant first patch. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix(ci): free disk space before Docker build in test-server-root-path The Dockerfile.non_root build ran out of disk on the CI runner. Remove Android SDK, .NET, Boost, and GHC toolchains (~12GB) to free space. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-09 11:46:23 -07:00
Yuneng Jiang	51af6fedb3	[Infra] Harden supply chain: remove unused scripts, add pip binary-only install ... Remove ci_cd/publish-proxy-extras.sh (dead, unreferenced PyPI publish script) and .pre-commit-config.yaml (pulls external repos from GitHub on git commit). Add --only-binary :all: to scripts/install.sh to prevent execution of malicious setup.py during pip install. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-02 14:13:57 -07:00
Yuneng Jiang	5f63873dca	[Infra] Pin all Docker build dependencies to exact versions ... Pin every dependency across all Docker builds so upgrades are intentional. Verified by building all 3 production images and diffing pip freeze against known-good v1.83.0-nightly baselines — zero version drift. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-01 00:05:39 -07:00
Ishaan Jaff	8e61b32b8e	[Staging] - Ishaan March 17th (#23903) ... * feat(xai): add grok-4.20 beta 2 models with pricing (#23900) Add three grok-4.20 beta 2 model variants from xAI: - grok-4.20-multi-agent-beta-0309 (reasoning + multi-agent) - grok-4.20-beta-0309-reasoning (reasoning) - grok-4.20-beta-0309-non-reasoning Pricing (from https://docs.x.ai/docs/models): - Input: $2.00/1M tokens ($0.20/1M cached) - Output: $6.00/1M tokens - Context: 2M tokens All variants support vision, function calling, tool choice, and web search. Closes LIT-2171 * docs: add Quick Install section for litellm --setup wizard (#23905) * docs: add Quick Install section for litellm --setup wizard * docs: clarify setup wizard is for local/beginner use * feat(setup): interactive setup wizard + install.sh (#23644) * feat(setup): add interactive setup wizard + install.sh Adds `litellm --setup` — a Claude Code-style TUI onboarding wizard that guides users through provider selection, API key entry, and proxy config generation, then optionally starts the proxy immediately. - litellm/setup_wizard.py: wizard with ASCII art, numbered provider menu (OpenAI, Anthropic, Azure, Gemini, Bedrock, Ollama), API key prompts, port/master-key config, and litellm_config.yaml generation - litellm/proxy/proxy_cli.py: adds --setup flag that invokes the wizard - scripts/install.sh: curl-installable script (detect OS/Python, pip install litellm[proxy], launch wizard) Usage: curl -fsSL https://raw.githubusercontent.com/BerriAI/litellm/main/scripts/install.sh | sh litellm --setup * fix(install.sh): remove orange color, add LITELLM_BRANCH env var for branch installs * fix(install.sh): install from git branch so --setup is available for QA * fix(install.sh): remove stale LITELLM_BRANCH reference that caused unbound variable error * fix(install.sh): force-reinstall from git to bypass cached PyPI version * fix(install.sh): show pip progress bar during install * fix(install.sh): always launch wizard via $PYTHON_BIN -m litellm, not PATH binary * fix(install.sh): use litellm.proxy.proxy_cli module (no __main__.py exists) * fix(install.sh): suppress RuntimeWarning from module invocation * fix(install.sh): use Python bin-dir litellm binary to avoid CWD sys.path shadowing * fix(install.sh): use sysconfig.get_path('scripts') to find pip-installed litellm binary * fix(install.sh): redirect stdin from /dev/tty on exec so wizard gets terminal, not exhausted pipe * fix(install.sh): warn about git clone duration, drop --no-cache-dir so re-runs are faster * feat(setup_wizard): arrow-key selector, updated model names * fix(setup_wizard): use sysconfig binary to start proxy, not python -m litellm * feat(setup_wizard): credential validation after key entry + clear next-steps after proxy start * style(install.sh): show git clone warning in blue * refactor(setup_wizard): class with static methods, use check_valid_key from litellm.utils * address greptile review: fix yaml escaping, port validation, display name collisions, tests - setup_wizard.py: add _yaml_escape() for safe YAML embedding of API keys - setup_wizard.py: add _styled_input() with readline ANSI ignore markers - setup_wizard.py: change DIVIDER to _divider() fn to avoid import-time color capture - setup_wizard.py: validate port range 1-65535, initialize before loop - setup_wizard.py: qualify azure display names (azure-gpt-4o) to avoid collision with openai - setup_wizard.py: work on env_copy in _build_config to avoid mutating caller's dict - setup_wizard.py: skip model_list entries for providers with no credentials - setup_wizard.py: prompt for azure deployment name - setup_wizard.py: wrap os.execlp in try/except with friendly fallback - setup_wizard.py: wrap config write in try/except OSError - setup_wizard.py: fix _validate_and_report to use two print lines (no \r overwrite) - setup_wizard.py: add .gitignore tip next to key storage notice - setup_wizard.py: fix run_setup_wizard() return type annotation to None - scripts/install.sh: drop pipefail (not supported by dash on Ubuntu when invoked as sh) - scripts/install.sh: use litellm[proxy] from PyPI (not hardcoded dev branch) - scripts/install.sh: guard /dev/tty read with -r check for Docker/CI compat - scripts/install.sh: remove --force-reinstall to avoid downgrading dependencies - tests/test_litellm/test_setup_wizard.py: 13 unit tests for _build_config and _yaml_escape * style: black format setup_wizard.py * fix: address remaining greptile issues - Windows compat, YAML quoting, credential flow - guard termios/tty imports with try/except ImportError for Windows compat - quote master_key as YAML double-quoted scalar (same as env vars) - remove unused port param from _build_config signature - _validate_and_report now returns the final key so re-entered creds are stored - add test for master_key YAML quoting * fix: add --port to suggested command, guard /dev/tty exec in install.sh * fix: quote api_base in YAML, skip azure if no deployment, only redraw on state change * fix: address greptile review comments - _yaml_escape: add control character escaping (\n, \r, \t) - test: fix tautological assertion in test_build_config_azure_no_deployment_skipped - test: add tests for control character escaping in _yaml_escape * feat(ui): remove Chat UI page link and banner from sidebar and playground (#23908) * feat(guardrails): MCPJWTSigner - built-in guardrail for zero trust MCP auth (#23897) * Allow pre_mcp_call guardrail hooks to mutate outbound MCP headers * Enhance MCPServerManager to support hook-modified arguments and extra headers. Update tests to validate argument mutation and header injection behavior, including warnings for OpenAPI-backed servers when headers are present. * Refactor MCPServerManager to raise HTTPException for extra headers in OpenAPI-backed servers. Update tests to reflect this change, ensuring proper exception handling instead of logging warnings. * Allow pre_mcp_call guardrail hooks to mutate outbound MCP headers * Enhance MCPServerManager to support hook-modified arguments and extra headers. Update tests to validate argument mutation and header injection behavior, including warnings for OpenAPI-backed servers when headers are present. * Refactor MCPServerManager to raise HTTPException for extra headers in OpenAPI-backed servers. Update tests to reflect this change, ensuring proper exception handling instead of logging warnings. * feat(guardrails): add MCPJWTSigner built-in guardrail for zero trust MCP auth Signs outbound MCP tool calls with a LiteLLM-issued RS256 JWT so MCP servers can trust a single signing authority instead of every upstream IdP. Enable in config.yaml: guardrails: - guardrail_name: mcp-jwt-signer litellm_params: guardrail: mcp_jwt_signer mode: pre_mcp_call default_on: true JWT carries sub (user_id), act.sub (team_id, RFC 8693), tool-level scope, iss, aud, iat/exp/nbf. RSA-2048 keypair auto-generated at startup unless MCP_JWT_SIGNING_KEY env var is set. Adds /.well-known/jwks.json endpoint and jwks_uri to /.well-known/openid-configuration so MCP servers can verify LiteLLM-issued tokens via OIDC discovery. * Update MCPServerManager to raise HTTPException with status code 400 for extra headers in OpenAPI-backed servers. Adjust tests to verify the correct status code and exception message. * fix: address P1 issues in MCPJWTSigner - OpenAPI servers: warn + skip header injection instead of 500 - JWKS Cache-Control: 5min for auto-generated keys, 1h for persistent - sub claim: fallback to apikey:{token_hash} for anonymous callers - ttl_seconds: validate > 0 at init time * docs: add MCP zero trust auth guide with architecture diagram * docs: add FastMCP JWT verification guide to zero trust doc * fix: address remaining Greptile review issues (round 2) - mcp_server_manager: warn when hook Authorization overwrites existing header - __init__: remove _mcp_jwt_signer_instance from __all__ (private internal) - discoverable_endpoints: copy dict instead of mutating in-place on OIDC augmentation - test docstring: reflect warn-and-continue behavior for OpenAPI servers - test: update scope assertions for least-privilege (no mcp:tools/list on tool-call JWTs) * fix: address Greptile round 3 feedback - initialize_guardrail: validate mode='pre_mcp_call' at init time — misconfigured mode silently bypasses JWT injection, which is a zero-trust bypass - _build_claims: remove duplicate inline 'import re' (module-level import already present) - _types.py: add TODO comment explaining jwt_claims is forward-compat plumbing for a follow-up PR that will forward upstream IdP claims into outbound MCP JWTs * feat(mcp_jwt_signer): add verify+re-sign, claim ops, two-token model, configurable scopes Addresses all missing pieces from the scoping doc review: FR-5 (Verify + re-sign): MCPJWTSigner now accepts access_token_discovery_uri and token_introspection_endpoint. When set, the incoming Bearer token is extracted from raw_headers (threaded through pre_call_tool_check), verified against the IdP's JWKS (JWT) or introspected (opaque), and only re-signed if valid. Falls back to user_api_key_dict.jwt_claims for LiteLLM JWT-auth mode. FR-12 (Configurable end-user identity mapping): end_user_claim_sources ordered list drives sub resolution — sources: token:<claim>, litellm:user_id, litellm:email, litellm:end_user_id, litellm:team_id. FR-13 (Claim operations): add_claims (insert-if-absent), set_claims (always override), remove_claims (delete) applied in that order. FR-14 (Two-token model): channel_token_audience + channel_token_ttl issue a second JWT injected as x-mcp-channel-token: Bearer <token>. FR-15 (Incoming claim validation): required_claims raises HTTP 403 when any listed claim is absent; optional_claims passes listed claims from verified token into the outbound JWT. FR-9 (Debug headers): debug_headers: true emits x-litellm-mcp-debug with kid, sub, iss, exp, scope. FR-10 (Configurable scopes): allowed_scopes replaces auto-generation. Also fixed: tool-call JWTs no longer grant mcp:tools/list (overpermission). P1 fixes: - proxy/utils.py: _convert_mcp_hook_response_to_kwargs merges rather than replaces extra_headers, preserving headers from prior guardrails. - mcp_server_manager.py: warns when hook injects Authorization alongside a server-configured authentication_token (previously silent). - mcp_server_manager.py: pre_call_tool_check now accepts raw_headers and extracts incoming_bearer_token so FR-5 verification has the raw token. - proxy/utils.py: remove stray inline import inspect inside loop (pre-existing lint error, now cleaned up). Tests: 43 passing (28 new tests covering all FR flags + P1 fixes). * feat(mcp_jwt_signer): add verify+re-sign, claim ops, two-token model, configurable scopes (core) Remaining files from the FR implementation: mcp_jwt_signer.py — full rewrite with all new params: FR-5: access_token_discovery_uri, token_introspection_endpoint, verify_issuer, verify_audience + _verify_incoming_jwt(), _introspect_opaque_token() FR-12: end_user_claim_sources ordered resolution chain FR-13: add_claims, set_claims, remove_claims FR-14: channel_token_audience, channel_token_ttl → x-mcp-channel-token FR-15: required_claims (raises 403), optional_claims (passthrough) FR-9: debug_headers → x-litellm-mcp-debug FR-10: allowed_scopes; tool-call JWTs no longer over-grant tools/list mcp_server_manager.py: - pre_call_tool_check gains raw_headers param to extract incoming_bearer_token - Silent Authorization override warning fixed: now fires when server has authentication_token AND hook injects Authorization tests/test_mcp_jwt_signer.py: 28 new tests covering all FR flags + P1 fixes (43 total, all passing) * fix(mcp_jwt_signer): address pre-landing review issues - Remove stale TODO comment on UserAPIKeyAuth.jwt_claims — the field is already populated and consumed by MCPJWTSigner in the same PR - Fix _get_oidc_discovery to only cache the OIDC discovery doc when jwks_uri is present; a malformed/empty doc now retries on the next request instead of being permanently cached until proxy restart - Add FR-5 test coverage for _fetch_jwks (cache hit/miss), _get_oidc_discovery (cache/no-cache on bad doc), _verify_incoming_jwt (valid token, expired token), _introspect_opaque_token (active, inactive, no endpoint), and the end-to-end 401 hook path — 53 tests total, all passing * docs(mcp_zero_trust): rewrite as use-case guide covering all new JWT signer features Add scenario-driven sections for each new config area: - Verify+re-sign with Okta/Azure AD (access_token_discovery_uri, end_user_claim_sources, token_introspection_endpoint) - Enforcing caller attributes with required_claims / optional_claims - Adding metadata via add_claims / set_claims / remove_claims - Two-token model for AWS Bedrock AgentCore Gateway (channel_token_audience / channel_token_ttl) - Controlling scopes with allowed_scopes - Debugging JWT rejections with debug_headers Update JWT claims table to reflect configurable sub (end_user_claim_sources) * fix(mcp_jwt_signer): wire all config.yaml params through initialize_guardrail The factory was only passing issuer/audience/ttl_seconds to MCPJWTSigner. All FR-5/9/10/12/13/14/15 params (access_token_discovery_uri, end_user_claim_sources, add/set/remove_claims, channel_token_audience, required/optional_claims, debug_headers, allowed_scopes, etc.) were silently dropped, making every advertised advanced feature non-functional when loaded from config.yaml. Add regression test that asserts every param is wired through correctly. * docs(mcp_zero_trust): add hero image * docs(mcp_zero_trust): apply Linear-style edits - Lead with the problem (unsigned direct calls bypass access controls) - Shorter statement section headers instead of question-form headers - Move diagram/OIDC discovery block after the reader is bought in - Add 'read further only if you need to' callout after basic setup - Two-token section now opens from the user problem not product jargon - Add concrete 403 error response example in required_claims section - Debug section opens from the symptom (MCP server returning 401) - Lowercase claims reference header for consistency * fix(mcp_jwt_signer): fix algorithm confusion attack + add OIDC discovery 24h TTL - Remove alg from unverified JWT header; use signing_jwk.algorithm_name from JWKS key instead. Reading alg from attacker-controlled headers enables alg:none / HS256 confusion attacks. - Add _oidc_discovery_fetched_at timestamp and _OIDC_DISCOVERY_TTL = 86400 (24h). Without a TTL the cached discovery doc never refreshes, so IdP key rotation is invisible. --------- Co-authored-by: Noah Nistler <60981020+noahnistler@users.noreply.github.com> * fix(ci): stabilize CI - formatting, type errors, test polling, security CVEs, router bug, batch resolution Fix 1: Run Black formatter on 35 files Fix 2: Fix MyPy type errors: - setup_wizard.py: add type annotation for 'selected' set variable - user_api_key_auth.py: remove redundant type annotation on jwt_claims reassignment Fix 3: Fix spend accuracy test burst 2 polling to wait for expected total spend instead of just 'any increase' from burst 2 Fix 4: Bump Next.js 16.1.6 -> 16.1.7 to fix CVE-2026-27978, CVE-2026-27979, CVE-2026-27980, CVE-2026-29057 Fix 5: Fix router _pre_call_checks model variable being overwritten inside loop, causing wrong model lookups on subsequent deployments. Use local _deployment_model variable instead. Fix 6: Add missing resolve_output_file_ids_to_unified call in batch retrieve non-terminal-to-terminal path (matching the terminal path behavior) Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * chore: regenerate poetry.lock to sync with pyproject.toml Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix: format merged files from main and regenerate poetry.lock Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix(mypy): annotate jwt_claims as Optional[dict] to fix type incompatibility Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix(ci): update router region test to use gpt-4.1-mini (fix flaky model lookup) Replace deprecated gpt-3.5-turbo-1106 with gpt-4.1-mini + mock_response in test_router_region_pre_call_check, following the same pattern used in commit 717d37cc5b for test_router_context_window_check_pre_call_check_out_group. Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * ci: retry flaky logging_testing (async event loop race condition) Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix(ci): aggregate all mock calls in langfuse e2e test to fix race condition The _verify_langfuse_call helper only inspected the last mock call (mock_post.call_args), but the Langfuse SDK may split trace-create and generation-create events across separate HTTP flush cycles. This caused an IndexError when the last call's batch contained only one event type. Fix: iterate over mock_post.call_args_list to collect batch items from ALL calls. Also add a safety assertion after filtering by trace_id and mark all langfuse e2e tests with @pytest.mark.flaky(retries=3) as an extra safety net for any residual timing issues. Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix(ci): black formatting + update OpenAPI compliance tests for spec changes - Apply Black 26.x formatting to litellm_logging.py (parenthesized style) - Update test_input_types_match_spec to follow $ref to InteractionsInput schema (Google updated their OpenAPI spec to use $ref instead of inline oneOf) - Update test_content_schema_uses_discriminator to handle discriminator without explicit mapping (Google removed the mapping key from Content discriminator) Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * revert: undo incorrect Black 26.x formatting on litellm_logging.py The file was correctly formatted for Black 23.12.1 (the version pinned in pyproject.toml). The previous commit applied Black 26.x formatting which was incompatible with the CI's Black version. Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> * fix(ci): deduplicate and sort langfuse batch events after aggregation The Langfuse SDK may send the same event (e.g., trace-create) in multiple flush cycles, causing duplicates when we aggregate from all mock calls. After filtering by trace_id, deduplicate by keeping only the first event of each type, then sort to ensure trace-create is at index 0 and generation-create at index 1. Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com> --------- Co-authored-by: Noah Nistler <60981020+noahnistler@users.noreply.github.com> Co-authored-by: Cursor Agent <cursoragent@cursor.com> Co-authored-by: Ishaan Jaff <ishaan-jaff@users.noreply.github.com>	2026-03-18 15:09:01 -07:00
Ishaan Jaff	1f412bc6d8	[Feat] Add Tool Policies for AI Gateway (#22732) ... * fix: fix ui render * fix: fix minor bugs * refactor: use prisma functions instead of raw sql (safer) * fix(add-new-tiles-to-tool-policies): allow developer to see what's available * feat: ensure tool allowlist runs correctly for tool names + mcp's * refactor: more ui improvements * feat: working key tool blocking * feat(tools): show tool logs * refactor: backend code improvements * refactor: improve log viewer for tools * fix: address PR review feedback for tool access control - Add missing blocked_tools column to root schema.prisma (schema drift) - Invalidate ToolPolicyRegistry after policy mutations so changes take effect immediately - Remove dead code: unused get_effective_policies, get_tool_policies_cached, and helpers Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: race condition in permission resolution and remove duplicate allowlist check - Use atomic update_many with object_permission_id=None to prevent concurrent requests from creating orphaned permission rows and losing tool blocks - Remove duplicate allowed_tools enforcement from guardrail (already enforced in auth layer via check_tools_allowlist) - Move inline uuid import to module level Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * update to account for userAgent * UI - Add ToolDetails * input/output policy * LiteLLM_PolicyAttachmentTable * LiteLLM_PolicyAttachmentTable * fix: add _enqueue_tool_registry_upsert * fix: tool mgmt endpoints * tool mgmt endpoints * Update tests/test_litellm/proxy/db/test_tool_registry_writer.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update tests/test_litellm/proxy/db/test_tool_registry_writer.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * Update tests/test_litellm/proxy/db/test_tool_registry_writer.py Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * fix: sync root schema.prisma and fix test_tool_registry_writer for input/output policy - Migrate root schema.prisma LiteLLM_ToolTable from call_policy to input_policy/output_policy, add missing user_agent and last_used_at columns (now consistent with litellm/proxy/schema.prisma and litellm-proxy-extras) - Fix SpendLogToolIndex comment across all three schema files - Fix all call_policy references in test_tool_registry_writer.py: swapped update_tool_policy arguments, wrong get_tools_by_names return type assertions, _mock_tool_row setting call_policy instead of input_policy Addresses Greptile review feedback on PR #22732. Made-with: Cursor --------- Co-authored-by: Krrish Dholakia <krrishdholakia@gmail.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>	2026-03-03 20:22:20 -08:00
Krish Dholakia	67f90254ed	feat(guardrails): team-based guardrail registration and approval workflow (#22459) ... * feat(guardrails): team-based guardrail registration and approval workflow Add team-based guardrail submission system where teams can register Generic Guardrail API guardrails for admin review. Includes: - POST /guardrails/register endpoint for team-scoped submissions - Admin review endpoints (list/get/approve/reject submissions) - Team Guardrails tab in the UI dashboard - extra_headers support for forwarding client headers to guardrail APIs - Prisma schema migration for status, submitted_at, reviewed_at fields - Documentation for team-based guardrails and static/dynamic headers Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(guardrails): address review feedback - SSRF, silent failure, redundant query - Validate api_base URL scheme (http/https only) and hostname in register_guardrail to prevent SSRF via team submissions - Return warning field in approve response when in-memory initialization fails so admins know the guardrail won't work until next sync cycle - Eliminate redundant DB query in list_guardrail_submissions by fetching all team guardrails once and deriving both filtered list and summary counts from the single result set Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix(guardrails): add pending_review status guard to reject endpoint Prevent rejecting already-active or already-rejected guardrails, which would create a DB/memory inconsistency (active in memory but rejected in DB). Now mirrors the approve endpoint's status check. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-02 22:06:49 -08:00
Krish Dholakia	12c4876891	Agents - assign tools (#22064) ... * feat(proxy): add max_iterations limiter for agent session loops (#22058) Adds a new proxy hook that enforces a per-session cap on the number of LLM calls an agentic loop can make. Callers send a session_id with each request, and the hook counts calls per session, returning 429 when the configured max_iterations limit is exceeded. - Uses Redis Lua script for atomic increment (multi-instance safe) - Falls back to in-memory cache when Redis unavailable - Follows parallel_request_limiter_v3 pattern - Configurable via key metadata: {"max_iterations": 25} - Session counters auto-expire via TTL (default 1hr) Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> * feat: add new code execution dataset * feat(agent_endpoints/): allow giving agents keys * fix: ui fixes * feat: allow assigning mcp servers to agents * fix: eliminate duplicate DB queries in MCP agent auth and N+1 in agent listing (#22110) - Extract _get_agent_object_permission helper so _get_allowed_mcp_servers_for_agent and _get_agent_tool_permissions_for_server share a single DB fetch instead of each independently querying the same agent row (was 1+N queries per MCP request) - Use include={"object_permission": True} on find_many in get_all_agents_from_db to eagerly load permissions in one query instead of N+1 - Use include={"object_permission": True} on create/update/find_unique in all agent CRUD operations, removing attach_object_permission_to_dict follow-up calls Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-25 11:44:30 -08:00
Ryan Crabbe	e0ddb2a525	fix: guard print_aggregate against empty latencies	2026-02-23 10:06:10 -08:00
Ryan Crabbe	95d9514054	fix: add auth headers and empty latencies guard to benchmark script	2026-02-23 09:58:40 -08:00
Ryan Crabbe	94b76ea9ad	feat: add network_mock transport for benchmarking proxy overhead without real API calls ... Intercepts at httpx transport layer so the full proxy path (auth, routing, OpenAI SDK, response transformation) is exercised with zero-latency responses. Activated via `litellm_settings: { network_mock: true }` in proxy config.	2026-02-21 17:52:39 -08:00
Alexsander Hamir	7f81dea8b3	Add custom auth header support and increase default prompt size to 100k chars (#19436)	2026-01-20 13:25:12 -08:00
Alexsander Hamir	270b41b0f4	Simplify file comments (#19382)	2026-01-19 17:01:38 -08:00
Alexsander Hamir	0cd7763d5f	Add health check scripts and parallel execution support (#19295) ... - Add health_check_client.py for monitoring model availability - Add health_check_client_README.md with usage documentation - Add health_check_requirements.txt for dependencies - Add run_parallel_health_checks.ps1 (PowerShell version) - Add run_parallel_health_checks.sh (Bash version) - Organize all scripts under scripts/health_check/ directory	2026-01-19 08:38:38 -08:00
Xiaohan Fu	07fe9e8604	implement failopen option default to True on grayswan guardrail (#18266) ... * implement failopen option default to True * introduce a config to set the timeout limit (default to 30)	2026-01-06 15:17:05 +05:30
Alexsander Hamir	b635f92d90	Add benchmark_proxy_vs_provider.py script to scripts directory with usage examples (#17889)	2025-12-12 11:26:34 -08:00
Alexsander Hamir	762b429d6c	enhance: create_litellm_branch tool to be more robust (#17874)	2025-12-12 05:35:50 -08:00
Cesar Garcia	a7ad8a36a4	chore: cleanup unused scripts and fix misplaced test file (#17611) ... Remove scripts/ directory containing unused development/debug scripts: - mock_ibm_guardrails_server.py - test_groq_streaming_issue.py (debug for #12660) - test_mock_ibm_guardrails.py - update_readme_providers_table.py Move misplaced test file to correct location: - test_litellm/ -> tests/test_litellm/ (from PR #17221)	2025-12-08 19:00:55 -08:00
Alexsander Hamir	c44e075b2d	feat: add script to create branches with litellm_ prefix (#17606) ... Add utility scripts to create branches with litellm_ prefix from contributor branches. This helps maintain consistent branch naming conventions for CI/CD. - scripts/create_litellm_branch.sh (Bash for macOS/Linux) - scripts/create_litellm_branch.ps1 (PowerShell for Windows) Usage: ./scripts/create_litellm_branch.sh [source_branch] [new_branch_name] ./scripts/create_litellm_branch.ps1 [source_branch] [new_branch_name] Features: - Auto-prefixes branch names with litellm_ - Handles existing branches gracefully - Validates branch names - Supports local and remote source branches	2025-12-06 10:41:39 -08:00
Rob Geada	d35d9008c9	Ensure detector-id is passed as header to IBM detector server (#16649)	2025-11-14 19:35:49 -08:00
Ishaan Jaff	0428229032	[Docs] readme fixes add supported providers (#16109) ... * add provider test * docs readme.md * docs providers * order providers * test_providers_alphabetically_ordered * docs endpoint * fix config * add ENDPOINT_COLUMNS * add provider endpoints * docs fix	2025-10-30 17:23:20 -07:00
Krish Dholakia	ddacaf6c32	(feat) Organizations: allow org admins to create teams on UI + (feat) IBM Guardrails (#15924) ... * fix(oldteams.tsx): allow org admin to create team on ui * fix(oldteams.tsx): show org admin a dropdown of allowed orgs for team creation * docs(access_control.md): cleanup doc * feat(ibm_guardrails/): initial commit adding support for ibm guardrails on litellm allows user to use self-hosted ibm guardrails * feat(ibm_detector.py): working detector * docs(ibm_guardrails.md): document new ibm guardrails * fix: fix linting errors	2025-10-25 11:13:39 -07:00
Cole McIntosh	000ecad4e2	Fix Groq streaming ASCII encoding issue ... Replace iter_lines()/aiter_lines() with iter_text()/aiter_text() using explicit UTF-8 encoding to handle non-ASCII characters like µ in streaming responses. - Added utf8_iter_lines() and utf8_aiter_lines() helper functions - Ensures proper UTF-8 decoding of streaming response content - Added comprehensive tests for Unicode character handling Fixes #12660	2025-08-16 08:32:22 -05:00