Will it run?

Pick a model and a card. You get every shipped quantization against every context length and KV cache dtype — each cell computed independently — rather than one yes-or-no that hides the settings it assumed.

Popular models

then choose a card on the model page
Qwen3-Coder-30B-A3B-InstructQwen3.6-27BQwen3.6-35B-A3BQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTPQwen3.8-27BQwen3.5-9Bgemma-4-26B-A4B-itgemma-4-12B-itDeepSeek-V4-Flashnemotron-3.5-asr-streaming-0.6bQwen3.5-4Bgemma-4-12B-it-qat-q4_0-unquantizedQwythos-9B-Claude-Mythos-5-1Mgemma-4-E4B-itQwen3-30B-A3B-Thinking-2507gemma-4-31B-itMuse-Glimmer-30BQwen3-4BQwen3-8Bgemma-4-26B-A4B-it-qat-q4_0-unquantizedQwen3.5-122B-A10BLaguna-XS-2.1DeepSeek-V4-Flash-0731Qwen3.5-0.8B

Or start from your hardware

Arc A310 4GB 4GBGeForce RTX 3050 6GBGeForce RTX 5050 8GBGeForce RTX 3080 10GBGeForce RTX 2080 Ti 11GBGeForce RTX 5070 12GBGeForce RTX 5060 Ti 16GBApple M3 Pro 18GBGeForce RTX 3080 Ti 20GBGeForce RTX 5090 D V2 24GBGeForce RTX 5090 D 32GBApple M5 Max 36GBApple M5 Pro 48GBApple M5 Max 64GBApple M3 Ultra 96GBApple M5 Max 128GBApple M2 Ultra 192GBApple M3 Ultra 256GBApple M3 Ultra 512GB