tencent/Hunyuan-7B-Instruct
Full fine-tune vs tencent/Hunyuan-7B-Pretrain. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 129/129, MLP 96/96, Attention 128/128
11 derivatives checked
- bartowski/tencent_Hunyuan-7B-Instruct-GGUF: Same weights, converted
- mradermacher/Hunyuan-7B-Instruct-i1-GGUF: Quantization
- marcelone/Hunyuan-7B-Instruct-GGUF: Same weights, converted
- gabriellarson/Hunyuan-7B-Instruct-GGUF: Same weights, converted
- yujiepan/hunyuan-dense-v1-tiny-random: Different architecture
- yujiepan/hunyuan-tiny-random: Different architecture
- DevQuasar/tencent.Hunyuan-7B-Instruct-GGUF: Same weights, converted
- tencent/Hunyuan-7B-Instruct-AWQ-Int4: Quantization
- mradermacher/Hunyuan-7B-Instruct-GGUF: Same weights, converted
- tencent/Hunyuan-7B-Instruct-GPTQ-Int4: Quantization
- tencent/Hunyuan-7B-Instruct-FP8: Quantization