github/OpenHands
1
0
Fork 0
mirror of https://github.com/OpenHands/OpenHands.git synced 2025-12-26 05:48:36 +08:00
Code Issues Packages Projects Releases Wiki Activity

Azure completion_tokens fix (take two) (#6975)

Browse Source
This commit is contained in:
Engel Nyst 2025-02-27 02:28:01 +01:00 committed by GitHub
parent 7ea418d020
commit 8b234ae57c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
openhands/llm/llm.py
View File

@ -137,6 +137,7 @@ class LLM(RetryMixin, DebugMixin):
# set up the completion function
kwargs: dict[str, Any] = {
'temperature': self.config.temperature,
'max_completion_tokens': self.config.max_output_tokens,
}
if (
self.config.model.lower() in REASONING_EFFORT_SUPPORTED_MODELS
@ -146,6 +147,10 @@ class LLM(RetryMixin, DebugMixin):
kwargs.pop(
'temperature'
) # temperature is not supported for reasoning models
# Azure issue: https://github.com/All-Hands-AI/OpenHands/issues/6777
if self.config.model.startswith('azure'):
kwargs['max_tokens'] = self.config.max_output_tokens
kwargs.pop('max_completion_tokens')
self._completion = partial(
litellm_completion,
@ -156,7 +161,6 @@ class LLM(RetryMixin, DebugMixin):
base_url=self.config.base_url,
api_version=self.config.api_version,
custom_llm_provider=self.config.custom_llm_provider,
max_completion_tokens=self.config.max_output_tokens,
timeout=self.config.timeout,
top_p=self.config.top_p,
drop_params=self.config.drop_params,