Browser GPU test20 seconds
See what AI your GPU can actually run.
Kasoku tells you which local AI models your graphics card can run. A 20-second test in this tab, then the exact install steps. No account.
When a test finishes, the card model, score, tier, browser and OS go to the public leaderboard. Nothing that identifies you. Privacy
Adapter
Looking for a GPU
Detecting
Five tiers, from potato to beast.
A card lands on the lower of its speed and its memory. Each tier adds models you can run today. How tiers work
Potato
Any card
Tiny models and simple tools. Still more than nothing.
- Gemma 3 1B
- Moondream 2
Starter
250+ and 3 GB+
Small chat models, speech to text and background removal.
- Llama 3.2 3B
- Phi-4-mini
Solid
900+ and 6 GB+
Everyday 3-4B chat models and quick image tools.
- Gemma 4 E4B
- Llama 3.1 8B
Strong
2,400+ and 8 GB+
7-8B chat models at reading speed and local image generation.
- FLUX.1 schnell
Beast
6,000+ and 16 GB+
Larger models, SDXL-class images and long context.
- gpt-oss 20B