Models
Pick a model to see how much memory it needs at each quantization and which devices can run it. Models are grouped by family. Already have a machine? Start from your hardware · How fit is calculated
Most popular
Llama 3.1 8B8B params · 128K ctx · popular on 20 devicesPhi-33.8B params · 4K ctx · popular on 16 devices
- Llama 3.2 3B3B params · 128K ctx · popular on 16 devices
- Llama 3.2 1B1B params · 128K ctx · popular on 16 devices
Gemma 3 1B1B params · 32K ctx · popular on 16 devices
Mistral Small 3 7B7B params · 33K ctx · popular on 15 devices- Gemma 3 4B4B params · 128K ctx · popular on 14 devices
Qwen 2.5 7B7B params · 128K ctx · popular on 13 devices
No models match. Try a different search or family.
Meta Llama
- Llama 3.2 1B1B params · 128K ctx
- Llama 3.2 3B3B params · 128K ctx
- Llama 2 7B7B params · 4K ctx
- Llama 3 8B8B params · 8K ctx
- Llama 3.1 8B8B params · 128K ctx
- Llama 2 13B13B params · 4K ctx
Llama 3.1 70B70B params · 128K ctx- Llama 2 70B70B params · 4K ctx
- Llama 3 70B70.6B params · 8K ctx
- Llama 3.1 405B405B params · 128K ctx
Llama 3.3 8B8B params · 8K ctx- Llama 3.3 70B70.6B params · 8K ctx
