cerebras/Qwen3-Coder-REAP-25B-A3B
LoRA-style vs Qwen/Qwen3-Coder-30B-A3B-Instruct. 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 0/193, MLP 1999/2074, Attention 0/192, lm_head 0/1
8 derivatives checked
- bartowski/cerebras_Qwen3-Coder-REAP-25B-A3B-GGUF: Same weights, converted
- mradermacher/Qwen3-Coder-REAP-25B-A3B-i1-GGUF: Quantization
- Em-80/Qwen3-coder-REAP-25B-A3B-Rust-GGUF: Quantization
- csabakecskemeti/cerebras.Qwen3-Coder-REAP-25B-A3B-GGUF: Same weights, converted
- mradermacher/Qwen3-Coder-REAP-25B-A3B-GGUF: Quantization
- noctrex/Qwen3-Coder-REAP-25B-A3B-MXFP4_MOE-GGUF: Quantization
- az13770129/Qwen3-Coder-REAP-25B-A3B-mlx-4Bit: Quantization
- nbeerbower/Hemlock-Qwen3-Coder-REAP-25B-A3B: LoRA-style