Beamivalice PRO

beamster

·

https://github.com/beamivalice/sushi

AI & ML interests

Sushi quant for Apple Silicon

Recent Activity

liked a model 4 days ago

MiniMaxAI/MiniMax-H3-Image

posted an update 8 days ago

Last Qwen3.8 Flash update with STQ 2.75bpw v2.0 pack. It moves away from hybrid affine/stq format to pure STQ with +15% quality to match oQ4e. 64GB users can enjoy this while waiting for Qwen 4! Require Sushi 1.5.1 ro run.

updated a model 8 days ago

beamster/Qwen3.8-Flash-Next-Sushi-4bpw

View all activity

Organizations

None yet

New activity in beamster/DeepSeek-V4.1-Flash-1017-Sushi-STQ-1.75bpw 13 days ago

Is sushi streaming weights from SSD like Strata?

#1 opened 13 days ago by

New activity in beamster/Qwen3.8-Flash-Next-Sushi-4bpw 29 days ago

Two EXL3 packs built on this pack's trunk (open converter, 4-bit and 5-bit)

#3 opened 29 days ago by

New activity in beamster/Qwen3.8-Flash-Next-Sushi-2.6bpw about 1 month ago

FP16 model

#1 opened about 1 month ago by

New activity in beamster/Qwen3.8-Flash-Next-Sushi-4bpw about 1 month ago

Trying it right now vs oQ4e

#2 opened about 1 month ago by

New activity in beamster/Qwen3.8-Flash-Next-Sushi-3bpw about 2 months ago

crazy!

#1 opened about 2 months ago by

New activity in turboderp/Qwen3.8-27B-exl3 3 months ago

Question about Calibration Traces

#3 opened 3 months ago by

New activity in turboderp/DeepSeek-V4-Flash-0731-exl3 3 months ago

Self-Correct version please

#2 opened 3 months ago by

New activity in deepsweet/Qwen3.6-35B-A3B-MLX-VL-oQ8-FP16 7 months ago

M1 Ultra 64GB FP16 improvements

#1 opened 7 months ago by

M1 Ultra 64GB FP16 improvements

#1 opened 7 months ago by