view post Post 6,867 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. See translation ❤️ 6 6 👍 1 1 🔥 1 1 + Reply
beamster/Qwen3.8-Flash-Next-Sushi-STQ-2.75bpw Image-Text-to-Text • 25B • Updated 5 days ago • 1.59k • 13
beamster/DeepSeek-V4.1-Flash-1017-Sushi-STQ-1.75bpw Image-Text-to-Text • 261B • Updated 7 days ago • 3.19k • 20
Sushi STQ models Collection State-of-the-art Sushi Trellis Quant (STQ) models that run with sushi engine, github.com/beamivalice/sushi. • 2 items • Updated 7 days ago
beamster/DeepSeek-V4.1-Flash-1017-Sushi-STQ-1.75bpw Image-Text-to-Text • 261B • Updated 7 days ago • 3.19k • 20
beamster/DeepSeek-V4.1-Flash-1017-Sushi-STQ-1.75bpw Image-Text-to-Text • 261B • Updated 7 days ago • 3.19k • 20
Sushi EXL3 models Collection EXL3-Affine hybrid hyper efficient models that run with Sushi engine, github.com/beamster/sushi • 7 items • Updated 7 days ago
Sushi STQ models Collection State-of-the-art Sushi Trellis Quant (STQ) models that run with sushi engine, github.com/beamivalice/sushi. • 2 items • Updated 7 days ago