model benchmarks
LLM benchmark scores
Independent intelligence, coding and agentic scores, plus output speed, latency, end-to-end response time and the API price per million tokens. Live data for 500+ models, sorted however you like.
hot
Highlights
Top 10: IntelligenceArtificial Analysis Intelligence Index · higher is better
Top 10: CodingCoding index · higher is better
Top 10: Output speedOutput tokens / second · higher is better
Top 10: Cheapest input$ per 1M input tokens · lower is better
idx
Leaderboard
| # | Model | Provider | Intelligence | Coding | Agentic | Speed t/s | Latency s | E2E s | $/M in | $/M out | Cost/task |
|---|---|---|---|---|---|---|---|---|---|---|---|
Loading live benchmark data... | |||||||||||
Benchmark data by Artificial Analysis. Intelligence, coding and agentic indices, speed, latency and cost-per-task are independent measurements; see their methodology for details. Live API prices for every model are on the main table.