What is WhichModel and how does it track LLM prices?

Open-source MCP server scraping 100+ models from 10+ providers every 4 hours, normalizing input/output/cached prices plus capabilities like context windows and tool calling.

Does WhichModel replace manual model selection for AI apps?

It lets agents query real-time for cheapest/best-fit models, e.g., "tool calling under $0.002/call" — no more hardcoding disasters.

Is WhichModel free and reliable for production?

Yes, MIT-licensed, cross-verified data; but scraping risks provider changes — use as agent tool, not sole source.

☁️ Cloud & Databases

LLM Pricing Nightmare: Open-Source Tracker That Scrapes 100+ Models Every 4 Hours

Imagine hardcoding a model, shipping your app, then eating 10x costs three months later. WhichModel's open-source fix scrapes pricing chaos every 4 hours—no more surprises.

theAIcatchup Apr 10, 2026 3 min read

Fluctuating line graph of LLM model prices over time with WhichModel dashboard overlay

⚡ Key Takeaways

LLM prices change weekly across 100+ models; most teams ignore it and overpay. 𝕏
WhichModel's MCP server lets AI agents query optimal models in real-time, every 4 hours updated. 𝕏
Price rarely matches quality — cheap models handle 80% tasks as well as premiums. 𝕏

Published by

theAIcatchup

Community-driven. Code-first.

#AI agents #AI model tracking #LLM pricing #MCP protocol #WhichModel #model tracking

Worth sharing?

Get the best Open Source stories of the week in your inbox — no noise, no spam.

Originally reported by Dev.to

⚡ Key Takeaways

The 60-Second TL;DR

theAIcatchup

Share this article

Worth sharing?

Related Stories

AI Agents Are Bleeding Cash on Overkill Models — WhichModel Fixes That Fast

10K Daily LLM Calls? This 20-Line Router Slashes $6K Monthly Bills

AI Agents: Stop Overpaying for Models Now

Traders, Beware: AI Agents Without Real-Time Stock Data Are Useless — MCP Changes That

Stay in the loop