Split the monolithic LiteLLM proxy into independently scalable Kubernetes components to allow separate horizontal scaling of the LLM data plane and management API surfaces
- Add DatabaseURLSettings pydantic-settings model that assembles DATABASE_URL (and optional DATABASE_URL_READ_REPLICA) from discrete DATABASE_* env vars before Prisma initializes, supporting both IAM token auth (minting short-lived RDS tokens) and password auth; replaces the CLI-only path that componentized entrypoints bypass
- Add gateway component (port 4000) that trims the proxy route table to the LLM data-plane surface (chat, embeddings, completions, audio, realtime, provider passthroughs, health/metrics) via an allowlist applied inside the lifespan context so plugin-registered routes are captured
- Add backend component (port 4001) that exposes the management/admin surface (keys, users, teams, orgs, spend analytics, model management, SSO, audit logs) with a complementary allowlist
- Add ui component — Next.js static export served by nginx (port 3000) with RSC payload routing, asset prefix aliasing, and SPA fallback for dashboard routes
- Add migrations component with dedicated Dockerfile that runs prisma migrate deploy via a Helm pre-install/pre-upgrade Job, eliminating per-pod schema contention on the Prisma advisory lock
- Add Helm chart (helm/litellm) with separate Deployments, Services, HPAs, and ConfigMap for each component; shared _helpers.tpl emits DATABASE_*, IAM_TOKEN_DB_AUTH, REDIS_*, and DISABLE_SCHEMA_UPDATE env vars from chart values; ingress template routes traffic to the correct component by path prefix
- Add comprehensive tests for DatabaseURLSettings covering IAM auth, password auth, read replica fallbacks, operator-pinned URL preservation, and percent-encoding; add coverage test asserting gateway + backend allowlist union equals the full proxy route set
- Add pydantic-settings>=2.14.1 as a proxy extra dependency and update liccheck allowlist
Co-authored-by: Yassin Kortam <yassinkortam@g.ucla.edu>
The mutation-test workflow timed out at the 350-minute job cap when
running whole-folder mutation against litellm/proxy/management_endpoints/
(~30 files, ~1.5 MB of source). Every mutant was running the full
test suite, and mutants were generated for lines no test covers — which
would survive regardless, just wasting compute.
mutmut 3.x's mutate_only_covered_lines setting runs the suite once up
front to compute coverage, then skips mutating uncovered lines. This
cuts the mutant count dramatically and is the right semantic for the
score (no test → no kill possible → uncountable). Per-mutant test
filtering by function name is already automatic in mutmut 3.x; no
external coverage step is needed.
* 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>
Our `uv.lock` already resolves jinja2 to 3.1.6, so Docker / CI installs
get that version. The `pyproject.toml` floor was lagging at 3.1.0,
which means downstream consumers using `--resolution=lowest-direct` or
older constraint files can land on 3.1.0-3.1.5 instead of the version
we actually test against.
Aligns the declared floor with the resolved version so external
installers see the same baseline our test matrix exercises.
`uv lock` diff is metadata-only (no resolved-version drift).
* feat(audio_transcription): add NVIDIA Riva STT provider
Adds nvidia_riva as a new audio transcription provider, supporting both
NVCF-hosted and self-hosted Riva ASR deployments via gRPC streaming.
- Auto-resamples input audio to 16 kHz mono LINEAR_PCM (soundfile + numpy,
audioread fallback) so callers can send any common format.
- Maps OpenAI params: language (en -> en-US), response_format (text/json/
verbose_json), timestamp_granularities=["word"] -> enable_word_time_offsets,
word offsets converted ms -> s for verbose_json.
- Auth: NVCF when nvcf_function_id is set (SSL on by default), self-hosted
otherwise (SSL off by default), with explicit use_ssl override.
- gRPC errors wrapped via NvidiaRivaException -> litellm exception classes.
- Optional deps gated behind [stt-nvidia-riva] extra (nvidia-riva-client,
soundfile, audioread, numpy).
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(nvidia_riva): address PR review feedback
- handler: forward call-level `timeout` to streaming_response_generator
(kwarg-detected via inspect for older riva-client compat) so a stalled
Riva server cannot block the caller indefinitely.
- audio_utils: spill bytes to a tempfile before audioread.audio_open;
most audioread backends (FFmpeg, GStreamer) require a real filesystem
path and previously raised TypeError on BytesIO, breaking the mp3/m4a
fallback path.
- audio_utils: prefer soxr / scipy.signal.resample_poly for resampling
(anti-aliased polyphase) when installed, falling back to linear only
as a last resort. Avoids aliasing on 44.1/48 kHz -> 16 kHz downsamples.
- transformation: bare `es` now maps to es-ES (Castilian) instead of
es-US, matching BCP-47 conventions.
Co-authored-by: Cursor <cursoragent@cursor.com>
* chore: trigger CI re-run [stabilize loop 1/3]
* Update litellm/llms/nvidia_riva/audio_transcription/transformation.py
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
* chore: trigger CI re-run [stabilize loop 1/3]
* fix code qa
* fix lint
* fix mypy
* fix mypy
* Fix NVIDIA Riva ASR service lookup
* Fix NVIDIA Riva transcription payload logging
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: oss-pr-review-agent-shin[bot] <281797381+oss-pr-review-agent-shin[bot]@users.noreply.github.com>
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
Removing the new check-dependency-floors.yml workflow. It only fires when
pyproject.toml changes, which is rare; for those PRs, a maintainer can
run the same check by hand with one command. Documented that command in
a pyproject.toml comment next to the deps.
Also adds the missing upper bound on importlib-metadata (>=8.0.0,<9.0)
for consistency with every other entry in the list.
CI matrix on Python 3.13 caught three floors that predate cp313 prebuilt
wheels and would force users into a Rust/C build:
- tiktoken: 0.7.0 -> 0.8.0 (cp313 wheels start at 0.8)
- tokenizers: 0.20.0 -> 0.21.0 (cp313 wheels start at 0.21; sdist's
pyproject.toml pre-0.21 is also malformed for modern build backends)
- pydantic: 2.5.0 -> 2.10.0 (pydantic-core cp313 wheels start at 2.27,
shipped with pydantic 2.10)
Verified locally on Python 3.10 and 3.13: install at lowest-direct +
import litellm + import every openai-namespace symbol the codebase uses
all pass.
The 12 core `[project.dependencies]` entries in pyproject.toml were exact
`==` pins, a side effect of the Poetry → uv migration. This forces every
downstream package that lists litellm as a dependency to downgrade common
runtime libraries (openai, pydantic, aiohttp, click, jsonschema, ...) to
the exact versions we ship. Customers have flagged this as a coexistence
blocker.
Switch to lower-bounded ranges with upper bounds where the upstream
package is pre-1.0 or has a known breaking-major-version policy.
Reproducibility for our Docker proxy and CI continues to come from
`uv.lock`, which is regenerated here as a metadata-only diff (no
resolved versions or hashes change).
Inspired by #26157 (which got stranded on `litellm_oss_staging_04_21_2026`
when the forward-merge to internal staging in #26216 was closed). Floors
in this PR are tighter than #26157's: they were validated by installing
litellm at `--resolution=lowest-direct` and importing the openai-namespace
symbols the codebase actually uses.
Floor highlights vs #26157:
- openai >= 2.20 (was 2.0) — Responses API symbols + `Omit` need a 2.x mid-range floor
- httpx >= 0.28, < 1.0 (was no upper) — pre-1.0
- importlib-metadata >= 8.0 (was 6.0) — stay in tested major
- tokenizers >= 0.20, < 1.0 (was 0.19, no upper) — pre-1.0
- aiohttp >= 3.10, < 4.0 (was no upper) — bound major
- pydantic >= 2.5, < 3.0 — kept
- All other floors: keep tested major, add upper bound
Adds a `check-dependency-floors.yml` GitHub Actions workflow that
installs litellm at `--resolution=lowest-direct` on Python 3.10 and 3.13
and import-checks every openai symbol the codebase uses, so a future
floor regression fails fast in CI rather than silently in the field.
Per Yuneng's feedback, use a single @pytest.mark.vcr marker so one record
sweep populates cassettes for every marked test across all providers,
instead of forcing each test to bind to a hard-coded cassette path.
Changes vs. the initial scaffolding:
- Add 'pytest-recording==0.13.4' on top of vcrpy. Adopt its layout:
cassettes live at 'cassettes/<test_module>/<test_name>.yaml', resolved
automatically. New tests just decorate with '@pytest.mark.vcr' — no
imports or path bookkeeping.
- Move the shared filter/match config into a 'vcr_config' fixture in
'tests/llm_translation/conftest.py' (consumed by pytest-recording for
every marked test in the dir). Drop the standalone 'vcr_config.py'.
- Bulk record / replay via the standard '--record-mode' CLI flag:
'make test-llm-translation-record' now sweeps every '@pytest.mark.vcr'
test under tests/llm_translation in one shot. Optional 'TARGET=' var
scopes to a single file.
- Move existing cassettes to the per-test paths and update the local
in-process Anthropic regenerator to write to the same paths.
- Refresh README + Makefile target docs to match the sweep workflow.
Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>
Live LLM e2e tests have been draining provider billing accounts and going
flaky on outages (LIT-2683). This change introduces vcrpy-backed cassette
replay so CI can exercise the same end-to-end LiteLLM transformation paths
without hitting the live provider:
- Add 'vcrpy==8.1.1' to the dev dependency group.
- New 'tests/llm_translation/vcr_config.py' centralises the VCR config:
filters auth/secret headers and per-request response headers, matches on
method+URI+body, and exposes 'LITELLM_VCR_RECORD_MODE' for re-recording.
- New 'tests/llm_translation/test_anthropic_completion_vcr.py' demonstrates
the pattern with one non-streaming and one streaming Anthropic test that
replay from cassettes shipped under 'cassettes/'.
- New 'tests/llm_translation/cassettes/_record_anthropic_fixtures.py' lets
contributors regenerate the canned Anthropic cassettes against a local
in-process mock (no API key required), and 'cassettes/README.md' documents
the full record/replay/refresh workflow.
- New 'make test-llm-translation-record FILE=...' Makefile target to refresh
cassettes against the live API.
Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>
- Dockerfile: pin the unscoped `brace-expansion@5.0.5` alongside
`@isaacs/brace-expansion@5.0.1`. The scoped package only has 5.0.0
and 5.0.1 published; CVE-2026-33750's fix (5.0.5) is on the unscoped
package which npm also vendors. The override loop now swaps both.
- Revert `black` 26.3.1 -> 24.10.0, `pytest` 9.0.3 -> 8.3.5, and
`pytest-asyncio` 1.3.0 -> 1.2.0. The major-version bumps cause CI
lint (black reformats hundreds of files) and code-quality
(liccheck.ini has no entry for the new versions) failures. Both
CVEs are dev-only; skipping leaves no runtime exposure.
* bump litellm-proxy-extras version to 0.4.67
* bump litellm-proxy-extras pin to 0.4.67 in litellm pyproject
* regenerate uv.lock for litellm-proxy-extras 0.4.67
* bump litellm-enterprise version to 0.1.38
* bump litellm-enterprise pin to 0.1.38 in litellm pyproject
* regenerate uv.lock for litellm-enterprise 0.1.38
All three dependency bumps in this PR resolve on Python 3.10, so there
is no need to jump the floor all the way to 3.11. Also restore the
py3.10-specific lunary==1.4.36 pin that was collapsed when the floor
was temporarily at 3.11.
Now that requires-python starts at 3.11, the "python_version >= '3.9'"
and ">= '3.10'" markers are unconditionally true, and the "< '3.10'"
entries for psycopg, Pillow, pyarrow, langchain, lunary, and pylint can
never resolve. Drop the dead markers and remove the unreachable pins so
the dependency list reflects what actually gets installed.
Bumps orjson, fastapi-sso, and python-multipart to their latest releases
in the proxy extra, and raises the project python floor to 3.11 so the
updated pins can resolve. CI already runs on 3.11 / 3.12 / 3.13 and the
Docker images ship python 3.13, so the floor change aligns the declared
support range with what is actually tested and shipped.
langgraph-prebuilt 1.0.9 imports ExecutionInfo and ServerInfo from
langgraph.runtime, but those symbols are not exported until
langgraph 1.1.0. Our pin of langgraph==1.0.10 allows
langgraph-prebuilt<1.1.0,>=1.0.8, and uv resolves to 1.0.9 (the
latest in range), which breaks at import time in every test that
touches langgraph.prebuilt (e.g. tests/pass_through_tests/test_mcp_routes.py):
ImportError: cannot import name 'ExecutionInfo' from 'langgraph.runtime'
Pinning langgraph-prebuilt to 1.0.8 pairs correctly with
langgraph==1.0.10 and restores the import path.
1. exclude-newer: change from absolute "2026-04-10" to relative "3 days".
All pinned deps were published before the 3-day cutoff. Re-locked so
uv lock --check passes in test-mcp.yml and test-linting.yml.
2. test_eager_tiktoken_load: run all 10 env var values in a single
subprocess instead of spawning 10 separate processes. Each cold
import litellm takes ~78s on CI, so the old loop took ~13 min on a
single xdist worker. Now takes ~78s total.
3. proxy-db remaining timeout: increase from 20 to 30 minutes. The
remaining group has 51 test files and was consistently timing out at
71% across all branches (pre-existing issue, not migration-related).
1. Cap requires-python to <3.14 — no deps ship 3.14 wheels yet, and
uv's cross-version resolver fails on the Python 3.14 split.
2. Change exclude-newer from relative "30 days" to absolute "2026-04-10"
so the lockfile stays reproducible. The relative date caused
cryptography==46.0.7 (published April 8) to fall outside the window.
3. Parametrize test_eager_loading_env_var_values instead of looping —
with xdist the 6 subprocess cases can run in parallel instead of all
running sequentially on one worker (~13 min → ~2 min).
Also removed redundant case variants (Yes/YES/On/ON) that test the
same str_to_bool code path.
* 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>
