armand0e/OmniCoder-9B-Claude-Opus-High-Reasoning-Distill
LoRA-style vs Tesslate/OmniCoder-9B. 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, out_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, Norms 0/215, Attention 56/332, MLP 96/208
4 derivatives checked
- mradermacher/OmniCoder-9B-Claude-Opus-High-Reasoning-Distill-i1-GGUF: Quantization
- mradermacher/OmniCoder-9B-Claude-Opus-High-Reasoning-Distill-GGUF: Same weights, converted
- armand0e/OmniCoder-9B-Claude-Opus-High-Reasoning-Distill-GGUF: Same weights, converted
- tianrui6641/omnicoder_local9b_blackwell_coregen_hdlfix_v2_hf_r64_epoch1-gguf: Quantization