yb2698 merges a week of LoRA adapters into one super-adapter, boosting Qwen coding scores by up to 20 points

yb2698 released an experimental "super-adapter" for Qwen3.8-27B: LoRA adapters trained over one week for different agent harnesses (toolchains) are concatenated along the rank dimension into one big LoRA, letting the model gain each harness's improvements without switching adapters. Measured results rose markedly; weights and training recipe are open-sourced, making this low-cost multi-task adapter merging approach worth watching.

Confirmed

Why It Matters

2026-11-20 ~ 2026-11-20 · 9 related posts

Full story(4 episodes)→

Primary sources

2 near-duplicate retellings: yb2698 · yb2698