N8Programs/Qwen3.6-35B-A3B-AntiLoop
LoRA-style vs Qwen/Qwen3.6-35B-A3B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, v_proj changed, in 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/5, Attention 190/392, MLP 0/399, Norms 0/247, lm_head 0/1, other 0/1
3 derivatives checked
- mradermacher/Qwen3.6-35B-A3B-AntiLoop-i1-GGUF: Quantization
- mradermacher/Qwen3.6-35B-A3B-AntiLoop-GGUF: Quantization
- mlx-community/Qwen3.6-35B-A3B-AntiLoop-NVFP4: Quantization