Status: [LIVE] Endpoints found: https://modelgrep.com/api/v1 https://modelgrep.com/api/v1/models?capabilities=vision&max_price=1&sort=price_input&limit=5 https://modelgrep.com/api/v1/rankings/coding/anthropic Source: https://github.com/sculptdotfun/modelgrep <div align="center"> <a href="https://modelgrep.com"><img src="docs/og.png" alt="modelgrep — find & understand every LLM" width="720"></a> # modelgrep **Find and understand every LLM.** Compare 300+ models by benchmark, speed, latency, price, context and capability — in one place. [](https://modelgrep.com) [](https://modelgrep.com/api) [](https://nextjs.org) [](https://www.typescriptlang.org) [](https://github.com/sculptdotfun/modelgrep/stargazers) [**modelgrep.com**](https://modelgrep.com) · [**API docs**](https://modelgrep.com/api) </div> modelgrep aggregates [OpenRouter](https://openrouter.ai)'s catalog with independent benchmarks, live performance and pricing so you can compare large language models by intelligence, speed, latency, cost, context and capability. --- ## What it does - **Leaderboard** — every model, filterable by capability (reasoning, tools, vision, JSON), price, context and provider, sortable on any metric. - **Rankings** — data-backed "best LLM for X" pages: smartest, coding, design, fastest, cheapest, free, reasoning, vision, agents, long-context, open-source — and each scoped per maker (e.g. *fastest Anthropic model*). - **Compare** — side-by-side model vs. model. - **Makers & model pages** — per-provider breakdowns with live throughput, latency, uptime and per-endpoint pricing. - **Free public API** — the same data as read-only JSON, no key required. ## Data sources | Source | Provides | | --- | --- | | OpenRouter | Base catalog, pricing, capabilities, providers, live thro