TeichAI/gemma-4-26B-A4B-it-Claude-Opus-Distill-v2
LoRA-style vs unsloth/gemma-4-26B-A4B-it. 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 30 of 30 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, MLP 90/231, Norms 0/433, other 0/122, Attention 115/223
6 derivatives checked
- TeichAI/gemma-4-26B-A4B-it-Claude-Opus-Distill-v2-GGUF: Same weights, converted
- mradermacher/gemma-4-26B-A4B-it-Claude-Opus-Distill-v2-i1-GGUF: Quantization
- mradermacher/gemma-4-26B-A4B-it-Claude-Opus-Distill-v2-GGUF: Quantization
- chfm/gemma-4-26B-A4B-it-Claude-Opus-Distill-v2-GGUF: Same weights, converted
- zecanard/gemma-4-26B-A4B-it-Claude-Opus-Distilled-v2-MLX-8bit-mxfp8: Quantization
- zecanard/gemma-4-26B-A4B-it-Claude-Opus-Distilled-v2-MLX-4bit-nvfp4: Quantization