[Bug]: When calling remote Ollama model, LiteLLM tries to access localhost for cost calculation

[FlorianVal]

What happened?

When Calling a remote Ollama server, LiteLLM makes a call to localhost in order to compute cost. I don't expect LiteLLM to call localhost if I'm running with non-localhost ollama server.

Related to #9602. When passing api_base to completion() it is not fully taken into account, Thus calling localhost.

Relevant log output

INFO:httpx:HTTP Request: POST http://my_remote_server:11434/api/generate "HTTP/1.1 200 OK"
19:42:06 - LiteLLM:INFO: utils.py:1211 - Wrapper: Completed Call, calling success_handler
INFO:LiteLLM:Wrapper: Completed Call, calling success_handler
19:42:06 - LiteLLM:INFO: cost_calculator.py:655 - selected model name for cost calculation: ollama/qwen3:0.6b
INFO:LiteLLM:selected model name for cost calculation: ollama/qwen3:0.6b
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"
19:42:06 - LiteLLM:INFO: cost_calculator.py:655 - selected model name for cost calculation: qwen3:0.6b
INFO:LiteLLM:selected model name for cost calculation: qwen3:0.6b
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"
19:42:06 - LiteLLM:INFO: cost_calculator.py:655 - selected model name for cost calculation: ollama/qwen3:0.6b
INFO:LiteLLM:selected model name for cost calculation: ollama/qwen3:0.6b
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"
19:42:06 - LiteLLM:INFO: cost_calculator.py:655 - selected model name for cost calculation: qwen3:0.6b
INFO:LiteLLM:selected model name for cost calculation: qwen3:0.6b
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"
INFO:httpx:HTTP Request: POST http://localhost:11434/api/show "HTTP/1.1 404 Not Found"

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

V1.70.0

Twitter / LinkedIn details

No response

[brian316]

yes this happened unbeknownst to us. made debugging challenging. code shouldn't do unexpected/unknown behavior

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[nathan-K-]

It's really annoying, can we reopen this issue ?