scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-10 22:41:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 83

fix(constants): make ANTHROPIC_MIN_THINKING_BUDGET_TOKENS a plain constant

Browse source 
The documentation CI test (tests/documentation_tests/test_env_keys.py)
asserts every os.getenv() key in the source has a matching entry in the
litellm-docs config_settings.md table. ANTHROPIC_MIN_THINKING_BUDGET_TOKENS
tracks Anthropic's published wire-protocol minimum (1024) — it's not a
user-tunable, so making it env-overridable was wrong anyway. Drop the
os.getenv() wrapper; the value is now a plain literal.

Co-authored-by: Mateo Wang <mateo-berri@users.noreply.github.com>
This commit is contained in:
Cursor Agent 2026-05-03 04:02:29 +00:00 • committed by mateo-berri
parent a6c673e7b9
commit 13714586b6
1 changed files with 6 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
litellm/constants.py
View file

@ -402,13 +402,12 @@ BEDROCK_MIN_THINKING_BUDGET_TOKENS = int(
# Anthropic's Messages API rejects ``thinking.budget_tokens < 1024`` with a
# 400. ``reasoning_effort='minimal'`` historically mapped to 128 (the global
# default) which always 400'd against direct Anthropic, Azure AI Anthropic,
# Vertex AI Anthropic, and Bedrock Invoke. Use the provider minimum so
# ``minimal`` is a usable tier on all Anthropic-backed routes; Bedrock
# Converse already clamped to 1024 server-side, so this just unifies the
# behavior.
ANTHROPIC_MIN_THINKING_BUDGET_TOKENS = int(
os.getenv("ANTHROPIC_MIN_THINKING_BUDGET_TOKENS", 1024)
)
# Vertex AI Anthropic, and Bedrock Invoke. Floor at the provider minimum so
# ``minimal`` is a usable tier on every Anthropic-backed route; Bedrock
# Converse already clamps server-side, this just unifies the behavior.
# Constant — not env-overridable — because it tracks Anthropic's published
# wire-protocol minimum, not a tunable.
ANTHROPIC_MIN_THINKING_BUDGET_TOKENS = 1024
REPLICATE_POLLING_DELAY_SECONDS = float(
os.getenv("REPLICATE_POLLING_DELAY_SECONDS", 0.5)
)