scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-11 22:51:28 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 154

fix: support response_text key for Cloudflare Workers AI Nemotron models

Browse source 
Fixes Issue #24065 - Cloudflare Workers AI API Error when using Nemotron 3 120B model.
Added fallback support to check for both response and response_text keys.
This commit is contained in:
RoomWithOutRoof 2026-03-19 08:39:45 +08:00
parent 488b93c477
commit d1fda5a2ad
1 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
litellm/llms/cloudflare/chat/transformation.py
View file

@ -147,9 +147,9 @@ class CloudflareChatConfig(BaseConfig):
) -> ModelResponse:
completion_response = raw_response.json()
model_response.choices[0].message.content = completion_response["result"][ # type: ignore
"response"
]
# Support both "response" and "response_text" keys (newer models like Nemotron use "response_text")
result = completion_response["result"]
model_response.choices[0].message.content = result.get("response") or result.get("response_text", "") # type: ignore
prompt_tokens = litellm.utils.get_token_count(messages=messages, model=model)
completion_tokens = len(