TeichAI/Nemotron-Cascade-14B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill
LoRA-style vs nvidia/Nemotron-Cascade-14B-Thinking. 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 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/161, Attention 160/160, MLP 120/120, lm_head 0/1, Embeddings 0/1
3 derivatives checked
- mradermacher/Nemotron-Cascade-14B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill-i1-GGUF: Quantization
- mradermacher/Nemotron-Cascade-14B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill-GGUF: Quantization
- TeichAI/Nemotron-Cascade-14B-Thinking-Claude-4.5-Opus-High-Reasoning-Distill-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.