Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses
stared · hn · 2026-09-08
- Quesma benchmarked multiple quantized versions of Qwen3.8 27B.
- Findings: 4-bit quantization performs on par with the full model and remains a practical choice, while 1-bit collapses badly and is unusable.
- A useful reference for anyone deploying mid-size models on local or low-cost hardware.
More from Infra
- M.2-Oculink eGPU Link Silently Downgrades to PCIe Gen1 — Here's How to Check — El_90 · 2026-09-11
- DeepSeek launches V4.1-Flash with 1M-token context and 4x smaller KV-cache — matlabulous · 2026-09-11
- What Can You Still Run on 8GB VRAM? User Asks for Small Models With Tool Use — riceinmybelly · 2026-09-11
- Spain's hourly 80% renewable matching rules clash as France fast-tracks 700MW sites, UK cuts grid queues — eherrerosj · 2026-09-11
- AI could add 0.3-0.4 points to Europe's productivity growth, but the EU holds under 5% of global compute — rohanpaul_ai · 2026-09-11
- Qualcomm's next-gen Hexagon NPU runs 30B MoE models with 32K context on-device — lee_stott · 2026-09-11