nomyo-ai/nomyo-router
2
2
Fork 0
Code Issues 1 Pull requests Projects Releases 5 Packages 1 Wiki Activity Actions

fix: params is never defined in ollama native backend

Browse source
This commit is contained in:
Alpha Nerd 2026-04-08 13:01:56 +02:00
parent f364a2d123
commit a432a65396
Signed by: alpha-nerd
SSH key fingerprint: SHA256:QkkAgVoYi9TQ0UKPkiKSfnerZy2h4qhi3SVPXJmBN+M
1 changed files with 5 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
router.py
View file

@ -2110,7 +2110,11 @@ async def chat_proxy(request: Request):
# Only cache when no max_tokens limit was set — otherwise # Only cache when no max_tokens limit was set — otherwise
# finish_reason=length might just mean max_tokens was hit, # finish_reason=length might just mean max_tokens was hit,
# not that the context window was exhausted. # not that the context window was exhausted.
_req_max_tok = params.get("max_tokens") or params.get("max_completion_tokens") or params.get("num_predict") _req_max_tok = (
params.get("max_tokens") or params.get("max_completion_tokens") or params.get("num_predict")
if use_openai else
(options.get("num_predict") if options else None)
)
if _dr == "length" and not _req_max_tok: if _dr == "length" and not _req_max_tok:
_pt = getattr(chunk, "prompt_eval_count", 0) or 0 _pt = getattr(chunk, "prompt_eval_count", 0) or 0
_ct = getattr(chunk, "eval_count", 0) or 0 _ct = getattr(chunk, "eval_count", 0) or 0