iapp/chinda-qwen3-4b
LoRA-style vs Qwen/Qwen3-4B. 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
- mradermacher/chinda-qwen3-4b-i1-GGUF: Quantization
- prithivMLmods/Chinda-Qwen3-4B-F32-GGUF: Same weights, converted
- DevQuasar/iapp.chinda-qwen3-4b-GGUF: Same weights, converted
- mradermacher/chinda-qwen3-4b-GGUF: Same weights, converted
- iapp/openthai-chinda-v1.0-qwen3-4b-gguf: Quantization