tencent/Hunyuan-4B-Instruct
Full fine-tune vs tencent/Hunyuan-4B-Pretrain. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 145/145, MLP 108/108, Attention 144/144
5 derivatives checked
- bartowski/tencent_Hunyuan-4B-Instruct-GGUF: Same weights, converted
- gabriellarson/Hunyuan-4B-Instruct-GGUF: Same weights, converted
- mradermacher/Hunyuan-4B-Instruct-i1-GGUF: Quantization
- mradermacher/Hunyuan-4B-Instruct-GGUF: Same weights, converted
- DevQuasar/tencent.Hunyuan-4B-Instruct-GGUF: Same weights, converted