{"slug": "unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up", "title": "Unsloth releases Dynamic v3.1 GGUFs for Qwen3.8-Flash-Next, says accuracy is up 15%", "summary": "Unsloth co-founder Daniel Han released Dynamic v3.1 GGUFs for Qwen3.8-Flash-Next on November 4th, claiming the quantized files score 10-15% higher than GSQ-RCO on DeepSWE, Terminal Bench, Aider Polyglot and other tests in Unsloth's own benchmarking, with accuracy up 15%. The release offers swappable engrams ranging from 28.8 GB to 102.4 GB, giving users a direct trade between file size and measured error for running the 125B-parameter model locally.", "body_md": "# Unsloth releases Dynamic v3.1 GGUFs for Qwen3.8-Flash-Next, says accuracy is up 15%\n\n**Daniel Han's Unsloth says the new quants beat GSQ-RCO by 10-15% on agentic coding tests, with swappable engrams from 28.8 GB to 102.4 GB.**\n\n        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)\n        · Published \n        · Updated \n\nPrimary source: [x.com](https://x.com/danielhanchen/status/2118037508139413742)\n\n## Why it matters\n\nQuantization affects whether a 125B-parameter model can run locally. Unsloth's engram switching gives users a direct trade between file size and measured error.\n\nUnsloth co-founder [Daniel Han (@danielhanchen)](https://x.com/danielhanchen) released new Dynamic v3.1 GGUFs for [Qwen3.8-Flash-Next](https://runtimewire.com/models/huggingface/qwen-qwen3.8-flash-next-c6e4ec527c4bd931) on November 4th, saying the files score 10-15% higher on DeepSWE, Terminal Bench, Aider Polyglot and other tests than GSQ-RCO in Unsloth's own benchmarking.…\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up", "canonical_source": "https://runtimewire.com/article/unsloth-dynamic-v31-ggufs-qwen38-flash-next", "published_at": "2026-11-04 17:59:04+00:00", "updated_at": "2026-11-04 21:19:25.271130+00:00", "lang": "en", "topics": ["large-language-models", "machine-learning", "ai-tools", "ai-infrastructure"], "entities": ["Unsloth", "Daniel Han", "Qwen3.8-Flash-Next", "Dynamic v3.1 GGUFs", "GSQ-RCO", "DeepSWE", "Terminal Bench", "Aider Polyglot"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up", "markdown": "https://wpnews.pro/news/unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up.md", "text": "https://wpnews.pro/news/unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up.txt", "jsonld": "https://wpnews.pro/news/unsloth-releases-dynamic-v3-1-ggufs-for-qwen3-8-flash-next-says-accuracy-is-up.jsonld"}}