Tensor-Level Allocation Boosts Qwen 3.5 4B Reasoning by 16.67%
devildip · reddit · 2026-08-22
The author successfully extended tensor-level allocation from the Gemma family to Qwen, achieving significant results on Qwen 3.5 4B.
Using the QLAB allocation strategy on IQ2XS quantization, reasoning performance improved from 46.875 to 54.688—a relative gain of 16.67%—with only a 0.412% increase in model size. The method involves building an imatrix from a category-based corpus, measuring damage, and redistributing precision at the tensor level without post-training or weight updates.
More from Models
- Nathan Lambert slams Meta's Muse Open License as 'trash' — natolambert · 2026-10-16
- Google launches Gemini 4 Argon to Ultra and Pro subscribers, touting physics and 3D skills — ShikharMurty · 2026-10-16
- Google gives X users early access to test Argon, not just companies — lyraxana · 2026-10-16
- Grok Voice 3.0 Tops Design Arena Speech Leaderboard, Up 7.4 Points — XFreeze · 2026-10-16
- Robotics LLMs tested on 16 tasks: GPT-6 Astra hits 17.2% success vs Fable's 9.1% — pabbeel · 2026-10-16
- Meta open-sources Muse Spark: top US open model, scores just below DeepSeek — bindureddy · 2026-10-16