bknyaz/Qwen3-Coder-Next-REAM
LoRA-style vs Qwen/Qwen3-Coder-Next. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/157, Attention 0/264, MLP 1742/1942, lm_head 0/1
6 derivatives checked
- mradermacher/Qwen3-Coder-Next-REAM-i1-GGUF: Quantization
- mradermacher/Qwen3-Coder-Next-REAM-GGUF: Full fine-tune
- TomLucidor/Qwen3-Coder-Next-REAM-mlx-3Bit: Quantization
- cyankiwi/Qwen3-Coder-Next-REAM-AWQ-4bit: Quantization
- TomLucidor/Qwen3-Coder-Next-REAM-mlx-4Bit: Quantization
- noctrex/Qwen3-Coder-Next-REAM-MXFP4_MOE-GGUF: Same weights, converted