hotdogs/qwen3.6-35b-opus-to-kimi-lora
Adapter vs lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled. hotdogs/qwen3.6-35b-opus-to-kimi-lora is a LoRA adapter, not full weights: rank 16, on k_proj, o_proj, q_proj, v_proj in 11 layers (44 adapted matrices). It modifies lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled at load time.