tiennm99/litellm
1
0
Fork 0
mirror of https://github.com/tiennm99/litellm.git synced 2026-08-17 08:25:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

[Feat] Day-0 support for GPT-5.5 and GPT-5.5 Pro (#26449)

Browse Source 
* feat(openai): day-0 support for GPT-5.5 and GPT-5.5 Pro

Add pricing + capability entries for the new GPT-5.5 family launched by
OpenAI on 2026-04-24:

- gpt-5.5 / gpt-5.5-2026-04-23 (chat): $5/$30/$0.50 per 1M
  input/output/cached input
- gpt-5.5-pro / gpt-5.5-pro-2026-04-23 (responses-only): $60/$360/$6
  per 1M input/output/cached input

Other fees (long-context >272k, flex, batches, priority, cache
discounts) follow the same ratios as GPT-5.4, with context window
retained at 1.05M input / 128K output.

No transformation / classifier code changes are required:
OpenAIGPT5Config.is_model_gpt_5_4_plus_model() already matches 5.5+ via
numeric version parsing, and model registration is driven from the
JSON. The existing responses-API bridge for tools + reasoning_effort
(litellm/main.py:970) already covers gpt-5.5-pro.

Tests:
- GPT5_MODELS regression list now covers gpt-5.5-pro and dated variants
- New test_generic_cost_per_token_gpt55_pro cost-calc test
- Updated test_generic_cost_per_token_gpt55 for long-context fields

* fix(openai): mirror reasoning_effort flags onto gpt-5.5 dated variants

gpt-5.5-2026-04-23 and gpt-5.5-pro-2026-04-23 were missing the
supports_none_reasoning_effort, supports_xhigh_reasoning_effort, and
supports_minimal_reasoning_effort flags that their non-dated
counterparts define. Reasoning-effort routing in OpenAIGPT5Config is
fully capability-driven from these JSON flags — since an absent flag
is treated as False for opt-in levels (xhigh), users pinning to a
dated snapshot would silently lose xhigh support and diverge from the
base alias on logprobs + flexible temperature handling.

Copy the flags onto both dated variants so every dated snapshot
inherits the base model's reasoning-effort capability profile.

Adds a parametrized regression test that asserts
supports_{none,minimal,xhigh}_reasoning_effort parity between each
dated variant and its non-dated counterpart, preventing future drift
when new snapshots are added.
This commit is contained in:
Mateo Wang
2026-04-24 14:10:42 -07:00
committed by GitHub
parent 7b47dffefd
commit d21e90f683
4 changed files with 382 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
litellm/model_prices_and_context_window_backup.json
+147 -1
View File
@@ -19275,13 +19275,24 @@
},
"gpt-5.5": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_flex": 2.5e-07,
"cache_read_input_token_cost_priority": 1e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_flex": 2.5e-06,
"input_cost_per_token_batches": 2.5e-06,
"input_cost_per_token_priority": 1e-05,
"litellm_provider": "openai",
"max_input_tokens": 272000,
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_flex": 1.5e-05,
"output_cost_per_token_batches": 1.5e-05,
"output_cost_per_token_priority": 6e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
@@ -19305,10 +19316,145 @@
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-2026-04-23": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_flex": 2.5e-07,
"cache_read_input_token_cost_priority": 1e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_flex": 2.5e-06,
"input_cost_per_token_batches": 2.5e-06,
"input_cost_per_token_priority": 1e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_flex": 1.5e-05,
"output_cost_per_token_batches": 1.5e-05,
"output_cost_per_token_priority": 6e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-pro": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"input_cost_per_token_flex": 3e-05,
"input_cost_per_token_batches": 3e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"output_cost_per_token_flex": 0.00018,
"output_cost_per_token_batches": 0.00018,
"supported_endpoints": [
"/v1/responses",
"/v1/batch"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-pro-2026-04-23": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"input_cost_per_token_flex": 3e-05,
"input_cost_per_token_batches": 3e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"output_cost_per_token_flex": 0.00018,
"output_cost_per_token_batches": 0.00018,
"supported_endpoints": [
"/v1/responses",
"/v1/batch"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.4": {
"cache_read_input_token_cost": 2.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 5e-07,
model_prices_and_context_window.json
+147 -1
View File
@@ -19289,13 +19289,24 @@
},
"gpt-5.5": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_flex": 2.5e-07,
"cache_read_input_token_cost_priority": 1e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_flex": 2.5e-06,
"input_cost_per_token_batches": 2.5e-06,
"input_cost_per_token_priority": 1e-05,
"litellm_provider": "openai",
"max_input_tokens": 272000,
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_flex": 1.5e-05,
"output_cost_per_token_batches": 1.5e-05,
"output_cost_per_token_priority": 6e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
@@ -19319,10 +19330,145 @@
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-2026-04-23": {
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"cache_read_input_token_cost_flex": 2.5e-07,
"cache_read_input_token_cost_priority": 1e-06,
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"input_cost_per_token_flex": 2.5e-06,
"input_cost_per_token_batches": 2.5e-06,
"input_cost_per_token_priority": 1e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"output_cost_per_token_flex": 1.5e-05,
"output_cost_per_token_batches": 1.5e-05,
"output_cost_per_token_priority": 6e-05,
"supported_endpoints": [
"/v1/chat/completions",
"/v1/batch",
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-pro": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"input_cost_per_token_flex": 3e-05,
"input_cost_per_token_batches": 3e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"output_cost_per_token_flex": 0.00018,
"output_cost_per_token_batches": 0.00018,
"supported_endpoints": [
"/v1/responses",
"/v1/batch"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.5-pro-2026-04-23": {
"cache_read_input_token_cost": 6e-06,
"cache_read_input_token_cost_above_272k_tokens": 1.2e-05,
"input_cost_per_token": 6e-05,
"input_cost_per_token_above_272k_tokens": 0.00012,
"input_cost_per_token_flex": 3e-05,
"input_cost_per_token_batches": 3e-05,
"litellm_provider": "openai",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"output_cost_per_token": 0.00036,
"output_cost_per_token_above_272k_tokens": 0.00054,
"output_cost_per_token_flex": 0.00018,
"output_cost_per_token_batches": 0.00018,
"supported_endpoints": [
"/v1/responses",
"/v1/batch"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": false,
"supports_system_messages": true,
"supports_tool_choice": true,
"supports_service_tier": true,
"supports_vision": true,
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
},
"gpt-5.4": {
"cache_read_input_token_cost": 2.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 5e-07,
tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py
+85 -1
View File
@@ -343,7 +343,10 @@ def test_generic_cost_per_token_gpt55():
assert model_cost_map["cache_read_input_token_cost"] == 5e-7
assert model_cost_map["litellm_provider"] == "openai"
assert model_cost_map["mode"] == "chat"
assert model_cost_map["max_input_tokens"] == 272000
# gpt-5.5 inherits GPT-5.4's long-context window + tiered pricing.
assert model_cost_map["max_input_tokens"] == 1050000
assert model_cost_map["input_cost_per_token_above_272k_tokens"] == 1e-5
assert model_cost_map["output_cost_per_token_above_272k_tokens"] == 4.5e-5
prompt_tokens = 1000
completion_tokens = 500
@@ -365,6 +368,87 @@ def test_generic_cost_per_token_gpt55():
)
def test_generic_cost_per_token_gpt55_pro():
"""gpt-5.5-pro: responses-only model — $60/1M input, $360/1M output, $6/1M cached input."""
model = "gpt-5.5-pro"
custom_llm_provider = "openai"
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
model_cost_map = litellm.model_cost[model]
# Sanity-check the map values match OpenAI's published pricing.
assert model_cost_map["input_cost_per_token"] == 6e-5
assert model_cost_map["output_cost_per_token"] == 3.6e-4
assert model_cost_map["cache_read_input_token_cost"] == 6e-6
assert model_cost_map["litellm_provider"] == "openai"
# gpt-5.5-pro is a responses-only model (no /v1/chat/completions endpoint).
assert model_cost_map["mode"] == "responses"
assert "/v1/chat/completions" not in model_cost_map["supported_endpoints"]
assert "/v1/responses" in model_cost_map["supported_endpoints"]
# Inherits GPT-5.4-pro's long-context window + tiered pricing (scaled 2x).
assert model_cost_map["max_input_tokens"] == 1050000
assert model_cost_map["input_cost_per_token_above_272k_tokens"] == 1.2e-4
assert model_cost_map["output_cost_per_token_above_272k_tokens"] == 5.4e-4
prompt_tokens = 1000
completion_tokens = 500
usage = Usage(
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
total_tokens=prompt_tokens + completion_tokens,
)
prompt_cost, completion_cost = generic_cost_per_token(
model=model,
usage=usage,
custom_llm_provider=custom_llm_provider,
)
assert round(prompt_cost, 10) == round(
model_cost_map["input_cost_per_token"] * prompt_tokens, 10
)
assert round(completion_cost, 10) == round(
model_cost_map["output_cost_per_token"] * completion_tokens, 10
)
@pytest.mark.parametrize(
"base_model,dated_model",
[
("gpt-5.5", "gpt-5.5-2026-04-23"),
("gpt-5.5-pro", "gpt-5.5-pro-2026-04-23"),
],
)
def test_gpt55_dated_variants_match_base_reasoning_effort_capabilities(
base_model, dated_model
):
"""Dated snapshots must carry the same reasoning_effort capability flags as
their non-dated counterparts.
Regression guard: ``supports_{none,minimal,xhigh}_reasoning_effort`` gate
downstream routing in ``OpenAIGPT5Config`` — a missing flag is treated as
``False`` for opt-in levels (e.g. ``xhigh``), which silently diverges
behavior between ``gpt-5.5`` and ``gpt-5.5-2026-04-23``. Pinning to a
dated variant must never lose capabilities relative to the base alias.
"""
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
base = litellm.model_cost[base_model]
dated = litellm.model_cost[dated_model]
for flag in (
"supports_none_reasoning_effort",
"supports_minimal_reasoning_effort",
"supports_xhigh_reasoning_effort",
):
assert dated.get(flag) == base.get(flag), (
f"{dated_model} has {flag}={dated.get(flag)!r}, "
f"but {base_model} has {flag}={base.get(flag)!r}. "
f"Dated snapshots must inherit the base model's reasoning_effort "
f"capability profile."
)
def test_generic_cost_per_token_anthropic_prompt_caching():
model = "claude-sonnet-4@20250514"
usage = Usage(
tests/test_litellm/llms/openai/test_is_model_gpt_5_model.py
+3
View File
@@ -47,6 +47,9 @@ GPT5_MODELS = [
"gpt-5.3",
"gpt-5.4",
"gpt-5.5",
"gpt-5.5-pro",
"gpt-5.5-2026-04-23", # dated variant
"gpt-5.5-pro-2026-04-23", # dated variant
"gpt-5.1-chat", # versioned chat — THE KEY REGRESSION CASE
"gpt-5.2-chat", # versioned chat — also a regression case
"gpt-5.3-chat", # versioned chat — THE KEY REGRESSION CASE