Compare model Elo scores, pricing, usage, speed, and benchmark results. Data may be live or simulated; source notes identify demo metrics.
AI Cap brings model rankings, API costs, usage, speed, and benchmark results into one comparison. Use the filters to narrow the list, compare model specifications, or estimate token costs for your workload.
Compare provider, context window, license, demo Elo score, input and output price per million tokens, reported usage, response speed, and available benchmark results. Filter open-weight and proprietary models or select models for a side-by-side comparison.
The Elo rankings and some benchmark values are simulated demo data, not a live voting leaderboard. When available, API pricing and usage are fetched from OpenRouter, speed data comes from llm-bench-data, and selected benchmark results are matched from Artificial Analysis data through ai-model-index. The status label and source notes show whether data is live, cached, or simulated.
Enter expected monthly input and output token volumes in the value calculator. It estimates cost from each model's per-million-token rates and compares estimated cost with the demo Elo score. Provider pricing can change, so verify current rates with the provider before budgeting.
There is no single best model for every use case. Compare quality indicators, benchmark coverage, context window, license, latency, and estimated cost against your requirements. Treat simulated scores as illustrative, not as a substitute for testing models on your own tasks.
0x03725858F0d8eCC50735463A224530B4B9b335F7Binance Smart Chain / EVM — send BNB, USDT, USDC, or any BEP-20 token. Always verify the network before sending.
Prices & usage: OpenRouter (CC BY 4.0). Speed: llm-bench-data (CC0).
Benchmarks: Artificial Analysis via ai-model-index · updated daily.
SWE-bench, Tau-bench, and battle responses are simulated for this demo.