AI models library optimized for on device.

Qwen models from Alibaba

Qwen 3.827BQwen 3.627BQwen 3.59BQwen 3.54BQwen 3.52BQwen 3.50.8B

Qwen 3.8 27B

2 available checkpoints

MediumQuantization106 tok/sSpeed

Measured on Apple M5 Max 128GB with speculative decoding

15.6 GBSize

LargeQuantization66 tok/sSpeed

Measured on Apple M5 Max 128GB with speculative decoding

28.8 GBSize

LFM models from LiquidAI

LFM 2.52.6BLFM 2.51.2B ThinkingLFM 2.51.2B InstructLFM 2.5350MLFM 2.5230M

LFM 2.5 2.6B

2 available checkpoints

MediumQuantization272 tok/sSpeed

Measured on Apple M5 Max 128GB

1.5 GBSize

LargeQuantization172 tok/sSpeed

Measured on Apple M5 Max 128GB

2.8 GBSize

Muse Glimmer models from Meta

Muse-Glimmer 30B

2 available checkpoints

MediumQuantization82 tok/sSpeed

Measured on Apple M5 Max 128GB with speculative decoding

16.6 GBSize

LargeQuantization52 tok/sSpeed

Measured on Apple M5 Max 128GB with speculative decoding

30.3 GBSize