nomyo-ai/nomyo-router
3
2
Fork 0
Code Issues 2 Pull requests 8 Projects Releases 10 Packages 1 Wiki Activity Actions

fix model naming to allow correct decrement usage counter in /v1 endpoints

Browse source
This commit is contained in:
Alpha Nerd 2025-11-24 09:33:54 +01:00
parent 7b50a5a299
commit 1c3f9a9dc4
1 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

8
router.py
View file

@ -1647,8 +1647,8 @@ async def openai_chat_completions_proxy(request: Request):
if prompt_tok != 0 or comp_tok != 0:
if not is_ext_openai_endpoint(endpoint):
if not ":" in model:
model = model+":latest"
await token_queue.put((endpoint, model, prompt_tok, comp_tok))
local_model = model+":latest"
await token_queue.put((endpoint, local_model, prompt_tok, comp_tok))
yield b"data: [DONE]\n\n"
else:
prompt_tok = async_gen.usage.prompt_tokens or 0
@ -1763,8 +1763,8 @@ async def openai_completions_proxy(request: Request):
if prompt_tok != 0 or comp_tok != 0:
if not is_ext_openai_endpoint(endpoint):
if not ":" in model:
model = model+":latest"
await token_queue.put((endpoint, model, prompt_tok, comp_tok))
local_model = model+":latest"
await token_queue.put((endpoint, local_model, prompt_tok, comp_tok))
# Final DONE event
yield b"data: [DONE]\n\n"
else: