zai-org/SWE-Dev-32B
Full fine-tune vs Qwen/Qwen2.5-Coder-32B-Instruct. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 127/129, MLP 192/192, Attention 448/448, lm_head 1/1
4 derivatives checked
- mradermacher/SWE-Dev-32B-i1-GGUF: Quantization
- csabakecskemeti/THUDM.SWE-Dev-32B-GGUF: Same weights, converted
- DevQuasar/zai-org.SWE-Dev-32B-GGUF: Same weights, converted
- mradermacher/SWE-Dev-32B-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.