Running 27B model on 12GB VRAM: Qwen 3.8 quantization benchmark
Square_Light1441 · reddit · 2026-08-28
A Reddit user shared a high-efficiency quantization setup for Qwen 3.8 27B, combining QAT Q2 weight quantization with Q5 KV Cache. The total RAM usage is around 13-14GB. Benchmarks show minimal performance degradation, allowing the model to run on a 12GB GPU (at 100K context) with capabilities claimed to surpass Claude Sonnet 4.6.
Related event: Qwen 3.8 27B Quantization Tests: 200K Context on 16GB VRAM(4 posts)→
More from Infra
- SK Hynix ties new US plant to looser chip equipment exports to China, sources say — pstAsiatech · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15
- vLLM-Omni ships Day-0 serving for FLUX 3 Image Klein with BF16, FP8, NVFP4 support — vllm_project · 2026-10-15
- NVIDIA's Nemotron 3.5 Super VL lands training and inference support on Applied Compute — ypatil125 · 2026-10-15
- CoreWeave bond yield tops 16% as loans outlast customer contracts by ~2 years — tengyanAI · 2026-10-15