Model costs endpoint broken down per provider

[toxic halo]

Hey! I wanted to reach out to ask about your https://openrouter.ai/api/v1/models endpoint, which we use for getting LLM costs. Do you have a way to see the costs broken down by provider? Something like Helicone has: https://www.helicone.ai/api/llm-costs

[strange pike]

https://openrouter.ai/docs/api-reference/list-endpoints-for-a-model

[toxic halo]

Thank you! I will check it out. At first glance, it looks like it’s fetching one model at a time. Do you have a way to fetch all the providers for all the models in one request?

[strange pike]

nope!

[toxic halo]

Would you consider showing the pricing per model on the models endpoint? We were using it at PostHog as the source of truth for LLM prices, but have had to switch to using Helicone’s endpoint, since we needed a breakdown per provider.

Given how many models/providers combinations there are, I’m not sure how realistic it is to make a call to that endpoints endpoint for every model is.

I’m asking because we’d love to switch back to using OpenRouter’s data for LLM pricing

[strange pike]

oh i'm not openrouter staff, i just answered your question because i'm been doing this lately while building my own tool for aggregating the data. there are currently 325 models/variants, so that many endpoint calls. https://orchid-three.vercel.app/endpoints

you'd probably get a quick response if you emailed [e-mail]

[vivid lotus]

so is the best way to do it just akin to getAllModels().map(modelName=>getEndpointsForModel(modelName)) ?

[strange pike]

that's what i do! you can actually do it pretty quickly