[Bug]: Mismatch in logic between v1/model/info endpoint and cost calculations #16884
Description
What happened?
Example:
When calling litellm with the gpt-51-chat model, cost tracking works fine. On litellm/llms/azure/cost_calculation.py:27, the get_model_info() function is called with model='azure/gpt-5.1-chat', which returns the correct costs.
But when you try to fetch the costs of the gpt-51-chat model through the v1/model/info endpoint, the same function is called (in get_litellm_model_info() at around litellm/proxy/proxy_server.py:3560) with the azure/ prefix removed. Since there is no entry in the cost mapping json for 'gpt-5.1-chat', the v1/model/info endpoint doesn't find any costs for the GPT 5.1 Chat model.
We haven't seen this issue with other models yet, but I suppose it's a broader issue that the logic isn't synced.
Relevant log output
Are you a ML Ops Team?
No
What LiteLLM version are you on ?
v1.77.5
Twitter / LinkedIn details
No response