Find the right leaderboard
Pick the kind of model, the task and your computer or budget. The leaderboard for that choice appears below the chart, and every leaderboard is listed after it.
Compare tested decision models and inference recipes across text, vision and workflow tasks. Use task filters and raw results: probability consistency measures coherence, not correctness.
Filter by your hardware and budget
detecting your OS…
↗ share
$ terminal
Score vs release date
x
show
● open weights ◆ API only ─┘ best score so far · x: release date
Release dates available for 95 of 357 models. Undated models are omitted; month-only dates are plotted at the start of the month.
$ your leaderboard
$ every leaderboard
By task
| task | any computer | 16GB Mac | 32GB Mac | 64GB Mac | 16GB graphics card | 32GB graphics card | 64GB graphics card |
|---|---|---|---|---|---|---|---|
| tools & automation | all · local | all | all | all | all · local | all · local | all · local |
| retrieval & classification | all · local | all | all | all | all · local | all · local | all · local |
| language understanding | all · local | all | all | all | all · local | all · local | all · local |
| arts & human taste | all · local | all | all | all | all · local | all · local | all · local |
| knowledge & reasoning | all · local | all | all | all | all · local | all · local | all · local |
| vision | all · local | all | all | all | all · local | all · local | all · local |
| workflows | all · local | all | all | all | all · local | all · local | all · local |
| consistency | all | all | all | all | all | all | all |
all: every model, including APIs · local: only models you can download and run
By computer
| Mac | 2GB · 4GB · 8GB · 16GB · 24GB · 32GB · 48GB · 64GB · 96GB · 128GB · 192GB · 256GB |
| PC without a graphics card | 2GB · 4GB · 8GB · 16GB · 24GB · 32GB · 48GB · 64GB · 96GB · 128GB · 192GB · 256GB |
| Graphics card | 1GB · 2GB · 4GB · 8GB · 12GB · 16GB · 24GB · 32GB · 48GB · 64GB · 80GB · 96GB · 128GB · 160GB |
| Graphics card, run locally | 1GB · 2GB · 4GB · 8GB · 12GB · 16GB · 24GB · 32GB · 48GB · 64GB · 80GB · 96GB · 128GB · 160GB |
By budget
run locally · under $0.05 per million input tokens · under $0.1 per million input tokens