scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-24 00:52:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 90

Merge pull request #42002 from BerriAI/litellm_cherrypick_1_98_x

Browse source 
fix(bedrock): backport #41870, the GPT-6 reasoning gate fix, and the Python 3.13 image pin to stable/1.98.x for v1.98.1
This commit is contained in:
Mateo Wang 2026-09-19 15:39:35 -07:00 • committed by GitHub
commit 86f64b0f88
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
11 changed files with 172 additions and 47 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
Dockerfile
View file

@ -1,10 +1,10 @@
# syntax=docker/dockerfile:1.7
# Base image for building
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
# Runtime image
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
ARG UI_BUILD_IMAGE=node:24.19-alpine3.24@sha256:d32cdf619f63fe0471182d08996dd516c6275bb5fd31ae06e55a570bd9e1ad43
@ -40,8 +40,8 @@ COPY --from=uvbin /uvx /usr/local/bin/uvx
RUN apk add --no-cache \
bash \
gcc \
python3 \
python3-dev \
python-3.13 \
python-3.13-dev \
rust \
openssl \
openssl-dev \
@ -51,6 +51,7 @@ RUN apk add --no-cache \
ENV UV_PROJECT_ENVIRONMENT=/app/.venv \
UV_LINK_MODE=copy \
UV_PYTHON_DOWNLOADS=0 \
PATH="/app/.venv/bin:${PATH}"
# Copy dependency metadata first for layer caching
@ -65,7 +66,7 @@ RUN uv sync --frozen --no-install-project --no-install-workspace --no-default-gr
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3
--python python3.13
# Copy full source tree
COPY . .
@ -86,7 +87,7 @@ RUN uv sync --frozen --no-default-groups --no-editable \
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3
--python python3.13
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
npm_config_cache=/root/.npm \
@ -101,7 +102,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
# node (without npm) is required by the prisma CLI at runtime
RUN apk add --no-cache bash openssl tzdata nodejs python3 libsndfile
RUN apk add --no-cache bash openssl tzdata nodejs python-3.13 libsndfile
WORKDIR /app
ENV PATH="/app/.venv/bin:${PATH}" \

12
backend/Dockerfile
View file

@ -1,5 +1,5 @@
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
FROM $UV_IMAGE AS uvbin
@ -16,7 +16,7 @@ COPY --from=uvbin /uv /uvx /usr/local/bin/
# instead of nodeenv downloading one whose dynamic deps may not be in Wolfi
# (e.g. Node 26.2.0 needs libatomic). Retry for transient apk.cgr.dev flakes.
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python3 python3-dev openssl openssl-dev libsndfile nodejs npm && break; \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev libsndfile nodejs npm && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -46,7 +46,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra proxy-runtime \
--extra extra_proxy \
--extra semantic-router \
--python python3
--python python3.13
# Stage 2 — copy source and install the project + workspace members.
COPY . .
@ -57,7 +57,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra proxy-runtime \
--extra extra_proxy \
--extra semantic-router \
--python python3
--python python3.13
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
npm_config_cache=/root/.npm \
@ -71,7 +71,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
RUN for i in 1 2 3; do \
apk add --no-cache bash openssl tzdata python3 libsndfile libatomic && break; \
apk add --no-cache bash openssl tzdata python-3.13 libsndfile libatomic && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done

15
docker/Dockerfile.database
View file

@ -1,10 +1,10 @@
# syntax=docker/dockerfile:1.7
# Base image for building
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
# Runtime image
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
ARG UI_BUILD_IMAGE=node:24.19-alpine3.24@sha256:d32cdf619f63fe0471182d08996dd516c6275bb5fd31ae06e55a570bd9e1ad43
@ -39,8 +39,8 @@ COPY --from=uvbin /uvx /usr/local/bin/uvx
RUN apk add --no-cache \
bash \
gcc \
python3 \
python3-dev \
python-3.13 \
python-3.13-dev \
openssl \
openssl-dev \
nodejs \
@ -49,6 +49,7 @@ RUN apk add --no-cache \
ENV UV_PROJECT_ENVIRONMENT=/app/.venv \
UV_LINK_MODE=copy \
UV_PYTHON_DOWNLOADS=0 \
PATH="/app/.venv/bin:${PATH}"
# Copy dependency metadata first for layer caching
@ -63,7 +64,7 @@ RUN uv sync --frozen --no-install-project --no-install-workspace --no-default-gr
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3
--python python3.13
# Copy full source tree
COPY . .
@ -84,7 +85,7 @@ RUN uv sync --frozen --no-default-groups --no-editable \
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3
--python python3.13
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
npm_config_cache=/root/.npm \
@ -98,7 +99,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
# node (without npm) is required by the prisma CLI at runtime
RUN apk add --no-cache bash openssl tzdata nodejs python3 libsndfile
RUN apk add --no-cache bash openssl tzdata nodejs python-3.13 libsndfile
WORKDIR /app
ENV PATH="/app/.venv/bin:${PATH}" \

17
docker/Dockerfile.non_root
View file

@ -1,8 +1,8 @@
# syntax=docker/dockerfile:1.7
# Base images
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG PROXY_EXTRAS_SOURCE=published
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
# Pinned by digest like the other base images; bump explicitly on Node upgrades.
@ -37,8 +37,8 @@ COPY --from=uvbin /uvx /usr/local/bin/uvx
RUN for i in 1 2 3; do \
apk add --no-cache \
python3 \
python3-dev \
python-3.13 \
python-3.13-dev \
gcc \
rust \
bash \
@ -52,6 +52,7 @@ RUN for i in 1 2 3; do \
ENV UV_PROJECT_ENVIRONMENT=/app/.venv \
UV_LINK_MODE=copy \
UV_PYTHON_DOWNLOADS=0 \
PATH="/app/.venv/bin:${PATH}" \
LITELLM_NON_ROOT=true \
XDG_CACHE_HOME=/app/.cache
@ -69,7 +70,7 @@ RUN --mount=type=cache,target=/app/.cache/uv,id=litellm-uv-cache \
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3
--python python3.13
# Copy full source tree
COPY . .
@ -96,7 +97,7 @@ RUN --mount=type=cache,target=/app/.cache/uv,id=litellm-uv-cache \
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3 \
--python python3.13 \
--no-sources-package litellm-proxy-extras; \
else \
uv sync --frozen --no-default-groups --no-editable \
@ -105,7 +106,7 @@ RUN --mount=type=cache,target=/app/.cache/uv,id=litellm-uv-cache \
--extra extra_proxy \
--extra semantic-router \
--extra saml \
--python python3; \
--python python3.13; \
fi
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
@ -124,7 +125,7 @@ RUN for i in 1 2 3; do \
apk upgrade --no-cache && break || sleep 5; \
done && \
for i in 1 2 3; do \
apk add --no-cache python3 bash openssl tzdata libsndfile nodejs && break || sleep 5; \
apk add --no-cache python-3.13 bash openssl tzdata libsndfile nodejs && break || sleep 5; \
done
# Copy only what runtime needs. The application is installed inside the venv;

12
gateway/Dockerfile
View file

@ -1,5 +1,5 @@
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
FROM $UV_IMAGE AS uvbin
@ -16,7 +16,7 @@ COPY --from=uvbin /uv /uvx /usr/local/bin/
# instead of nodeenv downloading one whose dynamic deps may not be in Wolfi
# (e.g. Node 26.2.0 needs libatomic). Retry for transient apk.cgr.dev flakes.
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python3 python3-dev openssl openssl-dev libsndfile nodejs npm && break; \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev libsndfile nodejs npm && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -47,7 +47,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra extra_proxy \
--extra semantic-router \
--extra bedrock-realtime \
--python python3
--python python3.13
# Stage 2 — copy source and install the project + workspace members.
COPY . .
@ -59,7 +59,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
--extra extra_proxy \
--extra semantic-router \
--extra bedrock-realtime \
--python python3
--python python3.13
RUN HOME=/opt/prisma XDG_CACHE_HOME=/opt/prisma/.cache PRISMA_BINARY_CACHE_DIR=/opt/prisma/binaries \
npm_config_cache=/root/.npm \
@ -73,7 +73,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
RUN for i in 1 2 3; do \
apk add --no-cache bash openssl tzdata python3 libsndfile libatomic && break; \
apk add --no-cache bash openssl tzdata python-3.13 libsndfile libatomic && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done

28
litellm/llms/bedrock/chat/converse_transformation.py
View file

@ -4,6 +4,7 @@ Translating between OpenAI's `/chat/completion` format and Amazon's `/converse`
import copy
import json
import re
import time
import types
from typing import Final, Literal, cast, overload
@ -92,6 +93,7 @@ BEDROCK_COMPUTER_USE_TOOLS: Final = [
"bash_",
"text_editor_",
]
BEDROCK_OPENAI_COMPAT_MIN_MAX_TOKENS: Final = 16
# Beta header patterns that are not supported by Bedrock Converse API
# These will be filtered out to prevent errors
@ -280,6 +282,14 @@ class AmazonConverseConfig(BaseConfig):
llm_provider="bedrock",
)
@staticmethod
def _requires_min_max_tokens(model: str) -> bool:
return re.search(r"openai\.gpt-\d|xai\.grok-", model) is not None
@staticmethod
def _is_openai_gpt_reasoning_model(model: str) -> bool:
return re.search(r"openai\.gpt-\d", model) is not None
def _is_nova_2_model(self, model: str) -> bool:
"""
Check if the model is a Nova 2 model that supports reasoningConfig.
@ -410,12 +420,16 @@ class AmazonConverseConfig(BaseConfig):
Handle the reasoning_effort parameter based on the model type.
- GPT-OSS models: passed through unchanged via additionalModelRequestFields.
- OpenAI GPT-5.x and GPT-6 models: mapped to ``reasoning.effort`` via additionalModelRequestFields.
- Nova 2 models: transformed to reasoningConfig.
- Anthropic models: mapped to ``thinking`` (and ``output_config.effort`` on
adaptive Claude 4.6 / 4.7).
"""
if "gpt-oss" in model:
optional_params["reasoning_effort"] = reasoning_effort
elif self._is_openai_gpt_reasoning_model(model):
reasoning: Final[BedrockConverseGptReasoningEffortBlock] = {"effort": reasoning_effort}
optional_params["reasoning"] = reasoning # rebind-ok: out-param store like siblings
elif self._is_nova_2_model(model):
reasoning_config: Final = self._transform_reasoning_effort_to_reasoning_config(reasoning_effort)
optional_params.update(reasoning_config)
@ -547,7 +561,11 @@ class AmazonConverseConfig(BaseConfig):
# only anthropic and mistral support tool choice config. otherwise (E.g. cohere) will fail the call - https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_ToolChoice.html
supported_params.append("tool_choice")
if "gpt-oss" in model:
if (
"gpt-oss" in model
or self._is_openai_gpt_reasoning_model(model)
or self._is_openai_gpt_reasoning_model(base_model)
):
supported_params.append("reasoning_effort")
elif self._is_nova_2_model(model):
# Nova 2 models support reasoning_effort (transformed to reasoningConfig)
@ -858,7 +876,11 @@ class AmazonConverseConfig(BaseConfig):
is_thinking_enabled=is_thinking_enabled,
)
if param == "max_tokens" or param == "max_completion_tokens":
optional_params["maxTokens"] = value
optional_params["maxTokens"] = (
max(value, BEDROCK_OPENAI_COMPAT_MIN_MAX_TOKENS)
if isinstance(value, int) and self._requires_min_max_tokens(model)
else value
)
if param == "stream":
optional_params["stream"] = value
if param == "stop":
@ -895,7 +917,7 @@ class AmazonConverseConfig(BaseConfig):
optional_params["_parallel_tool_use_config"] = {
"tool_choice": {"type": "auto", "disable_parallel_tool_use": not value}
}
if param == "thinking":
if param == "thinking" and not self._is_openai_gpt_reasoning_model(model):
if (
isinstance(value, dict)
and value.get("type") == "adaptive"

6
litellm/types/llms/bedrock.py
View file

@ -2,7 +2,7 @@ import json
from enum import Enum
from typing import TYPE_CHECKING, Any, Final, Literal
from typing_extensions import Required, TypedDict, override
from typing_extensions import ReadOnly, Required, TypedDict, override
from .openai import ChatCompletionToolCallChunk
@ -96,6 +96,10 @@ class BedrockConverseReasoningContentBlockDelta(TypedDict, total=False):
text: str
class BedrockConverseGptReasoningEffortBlock(TypedDict):
effort: ReadOnly[str]
class GuardrailConverseTextBlock(TypedDict, total=False):
text: str

12
migrations/Dockerfile
View file

@ -1,5 +1,5 @@
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:42df77a9974d6ec8b17a5ee8bc23b532600a44d705acef2409e0933c1251b45f
ARG LITELLM_BUILD_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG LITELLM_RUNTIME_IMAGE=cgr.dev/chainguard/wolfi-base@sha256:e624c5d5e42382ce7165ddafcbbf8e6769a24cbd02ea6114b880b05ae5ba2a8d
ARG UV_IMAGE=ghcr.io/astral-sh/uv:0.11.7@sha256:240fb85ab0f263ef12f492d8476aa3a2e4e1e333f7d67fbdd923d00a506a516a
FROM $UV_IMAGE AS uvbin
@ -35,7 +35,7 @@ COPY --from=uvbin /uv /uvx /usr/local/bin/
# instead of nodeenv downloading one whose dynamic deps may not be in Wolfi
# (e.g. Node 26.2.0 needs libatomic). Retry for transient apk.cgr.dev flakes.
RUN for i in 1 2 3; do \
apk add --no-cache bash gcc python3 python3-dev openssl openssl-dev libsndfile nodejs npm && break; \
apk add --no-cache bash gcc python-3.13 python-3.13-dev openssl openssl-dev libsndfile nodejs npm && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done
@ -56,7 +56,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
uv sync --frozen --no-install-project --no-install-workspace --no-default-groups --no-editable \
--extra proxy \
--extra extra_proxy \
--python python3
--python python3.13
# Stage 2 — copy source and install the project + workspace members.
COPY . .
@ -65,7 +65,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
uv sync --frozen --no-default-groups --no-editable \
--extra proxy \
--extra extra_proxy \
--python python3
--python python3.13
COPY migrations/run.py /app/run.py
@ -87,7 +87,7 @@ FROM $LITELLM_RUNTIME_IMAGE AS runtime
USER root
RUN for i in 1 2 3; do \
apk add --no-cache bash openssl tzdata python3 nodejs libsndfile libatomic && break; \
apk add --no-cache bash openssl tzdata python-3.13 nodejs libsndfile libatomic && break; \
[ $i = 3 ] && { echo "apk add failed after 3 retries" >&2; exit 1; }; \
sleep 5; \
done

4
pyproject.toml
View file

@ -1,6 +1,6 @@
[project]
name = "litellm"
version = "1.98.0"
version = "1.98.1"
description = "Library to easily interface with LLM API providers"
readme = "README.md"
requires-python = ">=3.10, <3.15"
@ -306,7 +306,7 @@ members = ["enterprise", "litellm-proxy-extras"]
profile = "black"
[tool.commitizen]
version = "1.98.0"
version = "1.98.1"
version_files = [
"pyproject.toml:^version",
]

96
tests/test_litellm/llms/bedrock/chat/test_converse_transformation.py
View file

@ -288,6 +288,102 @@ def test_reasoning_with_forced_tool_choice_switches_to_auto():
assert optional_params["tool_choice"] == {"auto": {}}
@pytest.mark.parametrize(
"model",
[
"us.openai.gpt-5.6-sol",
"global.openai.gpt-5.6-terra",
"bedrock/converse/us.openai.gpt-5.6-luna",
"us.openai.gpt-6-astra",
"bedrock/converse/global.openai.gpt-6-astra",
],
)
def test_reasoning_effort_maps_to_reasoning_effort_for_openai_gpt5_converse(model):
"""OpenAI GPT-5.x on Bedrock Converse routes reasoning_effort to
``additionalModelRequestFields.reasoning.effort`` rather than Anthropic ``thinking``."""
config = AmazonConverseConfig()
assert "reasoning_effort" in config.get_supported_openai_params(model=model)
optional_params = config.map_openai_params(
non_default_params={"reasoning_effort": "high"},
optional_params={},
model=model,
drop_params=False,
)
assert optional_params["reasoning"] == {"effort": "high"}
assert "thinking" not in optional_params
assert "reasoning_effort" not in optional_params
_, additional_request_params, _, _ = config._prepare_request_params(optional_params, model)
assert additional_request_params["reasoning"] == {"effort": "high"}
assert "thinking" not in additional_request_params
@pytest.mark.parametrize(
"model",
[
"us.openai.gpt-5.6-sol",
"bedrock/converse/global.openai.gpt-5.6-luna",
"us.openai.gpt-6-astra",
],
)
def test_openai_gpt5_converse_never_forwards_thinking(model):
"""GPT-5.x on Converse must never send Anthropic ``thinking``/``output_config`` (Bedrock rejects them).
Regression: ``thinking`` is not advertised as supported, and even when supplied alongside
``reasoning_effort`` in either order it never survives into the request."""
config = AmazonConverseConfig()
supported = config.get_supported_openai_params(model=model)
assert "thinking" not in supported
assert "output_config" not in supported
thinking_block = {"type": "enabled", "budget_tokens": 2048}
for non_default_params in (
{"reasoning_effort": "high", "thinking": thinking_block},
{"thinking": thinking_block, "reasoning_effort": "high"},
):
optional_params = config.map_openai_params(
non_default_params=dict(non_default_params),
optional_params={},
model=model,
drop_params=False,
)
_, additional_request_params, _, _ = config._prepare_request_params(optional_params, model)
assert additional_request_params["reasoning"] == {"effort": "high"}
assert "thinking" not in additional_request_params
@pytest.mark.parametrize(
"model, param, value, expected_max_tokens",
[
("us.openai.gpt-6-astra", "max_tokens", 1, 16),
("us.openai.gpt-6-astra", "max_completion_tokens", 1, 16),
("us.openai.gpt-6-astra", "max_tokens", 64, 64),
("us.xai.grok-4.6", "max_tokens", 1, 16),
("global.xai.grok-4.6", "max_completion_tokens", 1, 16),
("us.xai.grok-4.6", "max_tokens", 32, 32),
("anthropic.claude-sonnet-4-5-20250929-v1:0", "max_tokens", 1, 1),
("arn:aws:bedrock:us-east-1:123456789012:inference-profile/us.openai.gpt-6-astra", "max_tokens", 1, 16),
("arn:aws:bedrock:us-east-1:123456789012:inference-profile/global.xai.grok-4.6", "max_tokens", 1, 16),
("arn:aws:bedrock:us-east-1:123456789012:application-inference-profile/abc123xyz", "max_tokens", 1, 1),
],
)
def test_map_openai_params_enforces_minimum_max_tokens_for_openai_compat_models(
model: str, param: str, value: int, expected_max_tokens: int
):
optional_params = AmazonConverseConfig().map_openai_params(
non_default_params={param: value},
optional_params={},
model=model,
drop_params=False,
)
assert optional_params["maxTokens"] == expected_max_tokens
@pytest.mark.parametrize(
"model",
[

2
uv.lock generated
View file

@ -4194,7 +4194,7 @@ wheels = [
[[package]]
name = "litellm"
version = "1.98.0"
version = "1.98.1"
source = { editable = "." }
dependencies = [
{ name = "aiohttp" },