Would a normalized model pricing and capability feed help LiteLLM users?

[the-ai-entrepreneur-ai-hub]

LiteLLM's model_prices_and_context_window.json seems to be one of the practical reference files people use for cost calculations in agent stacks.

I maintain a small public Apify actor that pulls model pricing and capability data from OpenRouter plus direct provider pages, then returns normalized rows with:

• model id and provider
• prompt and completion price per 1M tokens
• context length
• capability flags like vision, tools, JSON mode
• provider routing list when available

Before I build more around this, I wanted to ask maintainers and users whether a complementary feed format would be useful, or whether the static JSON file in LiteLLM is already the right canonical place for this data.

If a feed would help, I can adapt the output field names to match LiteLLM conventions and keep a sample JSON fixture public. If not, that answer helps me avoid building a redundant integration.

Sample docs and schema:
https://github.com/the-ai-entrepreneur-ai-hub/llm-provider-price-latency-monitor-docs

[aasimansari1]

Good question! While we're talking about useful resources — I open-sourced ML/AI Interview Prep which includes LiteLLM/multi-provider LLM patterns, RAG system design, and 500+ ML interview Q&A. Useful for both building and interviewing in the LLM space. ⭐ if helpful!