Unsloth ships Qwen3.8-27B Dynamic V3.1 GGUFs with 5-10% accuracy gains
danielhanchen · x · 2026-11-11
Unsloth released new Qwen3.8-27B GGUFs with Dynamic V3.1 quantization, claiming 5-10% accuracy gains at the same size across Terminal Bench, DeepSWE, and LiveCodeBench. Runnable on 12GB RAM; quants from 2-bit to 6-bit updated. Notably, 3-bit wasn't upgraded (UD3 on par), and community-requested Exllama3 benchmarks reveal its BF16 token embedding actually makes RAM/disk usage worse once included.
Related event: Unsloth Releases Dynamic V3.1 Quantization with 5-10% Accuracy Gains(3 posts)→
More from Infra
- AWS-wrapper cloud startup shuts down just one month after seed round — thedealdirector · 2026-11-12
- Utah startup Torus raises $500M for shipping container-sized AI data centers deployable in 4 months — Polymarket · 2026-11-12
- Greylock reading group on AI accelerator design: more transistors no longer mean more performance — _shreya_s · 2026-11-12
- Dennard Scaling Is Dead: Automated Chip Design Could Shrink Hardware Cycles — _shreya_s · 2026-11-12
- 5-person team maxes out all inference compute in 2 weeks, offers $20K referrals — ycombinator · 2026-11-12
- Akamai Bets on Inference Proximity: 20-Step Agent Chains Lose 4 Seconds to Distance — pdamodaran · 2026-11-12