[Bug]: input_cost_per_token does not work

[ishaan-jaff]

What happened?

Tried this config, following this doc: https://docs.litellm.ai/docs/proxy/custom_pricing#cost-per-token-eg-azure

model_list:
  - model_name: expensive-openai-endpoint
    litellm_params:
      model: openai/fake
      api_key: fake-key
      api_base: https://exampleopenaiendpoint-production.up.railway.app/
      input_cost_per_token: 5
      output_cost_per_token: 5


general_settings:
  store_prompts_in_spend_logs: true

litellm_settings:
  callbacks: ["prometheus"]

With this request

curl -i http://0.0.0.0:4000/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-1234" \
-d '{
            "model": "expensive-openai-endpoint",
            "messages": [
                {
                    "role": "user",
                    "content": [
                                {"type": "text", "text": "hi"}
                    ]
                }
            ]
        }'

I see this in response headers, without any cost

date: Thu, 23 Jan 2025 19:45:19 GMT
server: uvicorn
content-length: 489
content-type: application/json
x-litellm-call-id: ffe79d11-ef12-4185-a51e-fc067f3ca975
x-litellm-model-id: c019fd0e2ba1a24b2231c45a6e404fffd6afbeb3b6259b4ddba3b70854fad952
x-litellm-model-api-base: https://exampleopenaiendpoint-production.up.railway.app/
x-litellm-version: 1.56.9
x-litellm-key-spend: 0.0
x-litellm-response-duration-ms: 254.09199999999998
x-litellm-overhead-duration-ms: 24.196
llm_provider-content-length: 384
llm_provider-content-type: application/json
llm_provider-date: Thu, 23 Jan 2025 19:45:18 GMT
llm_provider-server: railway-edge
llm_provider-x-railway-request-id: ICSDtxoeShGVwcslDpjkqw_3485859946
x-litellm-model-group: expensive-openai-endpoint
x-litellm-attempted-retries: 0

Relevant log output

-

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

1.59.5

Twitter / LinkedIn details

No response