Fimeg/letta-server
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

fix: correct ChatGPT OAuth GPT-5 max output token defaults (#9592)

Browse Source 
fix: align ChatGPT OAuth GPT-5 max output token defaults

Update ChatGPT OAuth provider defaults so GPT-5 family models report 128k max output tokens based on current OpenAI model docs, avoiding incorrect 16k values in /v1/models responses.

👾 Generated with [Letta Code](https://letta.com)

Co-authored-by: Letta <noreply@letta.com>
This commit is contained in:
jnjpng
2026-02-20 14:52:48 -08:00
committed by Caren Thomas
parent 73c824f5d2
commit 9c8589a687
1 changed files with 11 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
letta/schemas/providers/chatgpt_oauth.py
View File

@@ -310,15 +310,21 @@ class ChatGPTOAuthProvider(Provider):
) )
def get_default_max_output_tokens(self, model_name: str) -> int: def get_default_max_output_tokens(self, model_name: str) -> int:
"""Get the default max output tokens for ChatGPT models.""" """Get the default max output tokens for ChatGPT models.
References:
- https://developers.openai.com/api/docs/models/gpt-5
- https://developers.openai.com/api/docs/models/gpt-5-codex
- https://developers.openai.com/api/docs/models/gpt-5.1-codex-max
"""
# GPT-5 family (gpt-5, gpt-5.x, codex variants): 128k max output tokens
if "gpt-5" in model_name:
return 128000
# Reasoning models (o-series) have higher limits # Reasoning models (o-series) have higher limits
if model_name.startswith("o1") or model_name.startswith("o3") or model_name.startswith("o4"): if model_name.startswith("o1") or model_name.startswith("o3") or model_name.startswith("o4"):
return 100000 return 100000
# GPT-5.x models
elif "gpt-5" in model_name:
return 16384
# GPT-4 models # GPT-4 models
elif "gpt-4" in model_name: if "gpt-4" in model_name:
return 16384 return 16384
return 4096 return 4096