TheDrummer/Behemoth-X-123B-v2
LoRA-style vs mistralai/Mistral-Large-Instruct-2411. 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, q_proj, up_proj, v_proj changed, in 88 of 88 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/177, MLP 264/264, Attention 352/352, lm_head 0/1
3 derivatives checked
- bartowski/TheDrummer_Behemoth-X-123B-v2-GGUF: Quantization
- mratsim/Behemoth-X-123B-v2-NVFP4: Quantization
- otherwhere1/Behemoth-X-123B-v2-mlx-4Bit: Quantization
2 other repos hold byte-for-byte identical weight files.