Qwen3.8-27B-Ridge-3.7bpw released, shrinking model size to 11.7GB
udmrzn · x · 2026-08-17
Released Qwen3.8-27B-Ridge-3.7bpw, compressing the model size to 11.7 GiB. It features a GDN-aware quantization strategy where state tensors remain Q80 and Gated-DeltaNet mixers use Q4K, preserving native MTP and the full vision tower. It is now supported on llama.cpp, Ollama, and LM Studio.
More from Infra
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- Oracle down 12% in a week, nearly 50% off June high as OpenAI exposure spooks market — alvelda · 2026-10-16
- US data center moratoriums top 400 — an AI bull explains why they're actually good for AI — kevinsxu · 2026-10-16
- Microsoft Research unveils Magnetar, the fastest zkVM at 10-39x speedups and post-quantum security — jedisct1 · 2026-10-16
- Bonsai 2 2B hits 120 tok/s on iPhone 18 Pro, outpaces Qwen 3.5 2B at half the memory — cephaloform · 2026-10-16
- Samsung cuts Q4 phone output as Nittobo shifts E-glass to AI-chip T-glass — zephyr_z9 · 2026-10-16