SamsungSAILMontreal/Qwen3-30B-A3B-Instruct-2507-REAM
LoRA-style vs Qwen/Qwen3-30B-A3B-Instruct-2507. 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 1874/1922, Attention 0/192, lm_head 0/1
2 derivatives checked
- Akicou/Qwen3-30B-A3B-Instruct-2507-REAM-GGUF: Quantization
- sasa2000/Qwen3-30B-A3B-Instruct-2507-REAM-heretic: Abliterated