TeichAI/Qwen3-4B-Thinking-2507-Gemini-2.5-Flash-Distill
LoRA-style vs unsloth/Qwen3-4B-Thinking-2507. 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 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 108/108, Attention 144/144
5 derivatives checked
- DavidAU/Qwen3-4B-Gemini-TripleX-High-Reasoning-Thinking-Heretic-Uncensored-GGUF: Full fine-tune
- TeichAI/Qwen3-4B-Thinking-2507-Gemini-2.5-Flash-Distill-GGUF: Same weights, converted
- mradermacher/Qwen3-4B-Thinking-2507-Gemini-2.5-Flash-Distill-i1-GGUF: Quantization
- mradermacher/Qwen3-4B-Thinking-2507-Gemini-2.5-Flash-Distill-GGUF: Same weights, converted
- Richard944/Qwen3-4B-Gemini-TripleX-High-Reasoning-Thinking-Heretic-Uncensored-GGUF: Full fine-tune