Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled
LoRA-style vs Qwen/Qwen3.5-27B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/5, MLP 192/307, Attention 112/560, other 0/1, Norms 0/325
16 derivatives checked
- mradermacher/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-i1-GGUF: Quantization
- mlx-community/Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-4bit: Quantization
- mradermacher/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-GGUF: Quantization
- mlx-community/Huihui-Qwen3.5-27B-Claude-4.6-Opus-abliterated-4bit: Quantization
- mconcat/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-FP8-Dynamic: Quantization
- mlx-community/Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-6bit: Quantization
- huihui-ai/Huihui-Qwen3.5-27B-Claude-4.6-Opus-abliterated: Abliterated
- codgician/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-GPTQ-int4: Quantization
- dlindsey/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-i1-GGUF: Quantization
- mconcat/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-NVFP4: Quantization
- nightmedia/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-qx64-hi-mlx: Quantization
- Jackrong/MLX-Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-4bit: Quantization
- mlx-community/Huihui-Qwen3.5-27B-Claude-4.6-Opus-abliterated-6bit: Quantization
- peterjohannmedina/Medina-Qwen3.5-27B-OpenClaw: Adapter
- Smilyai-labs/Mira-1-XL: Full fine-tune
- SeaWolf-AI/Darwin-Qwen3.5-27B-x-Qwen3.5-27B-Claude-4-08162: Partial fine-tune
120 other repos hold byte-for-byte identical weight files.