zai-org/AutoGLM-Phone-9B
Full fine-tune vs zai-org/GLM-4.1V-9B-Base. Fine-tune: 91% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 158/213, MLP 80/155, Attention 280/328, lm_head 1/1, Embeddings 1/4, other 0/3
5 derivatives checked
- mradermacher/AutoGLM-Phone-9B-i1-GGUF: Quantization
- ggml-org/AutoGLM-Phone-9B-GGUF: Full fine-tune
- BlcaCola/AutoGLM-Phone-9B-GGUF: Partial fine-tune
- mradermacher/AutoGLM-Phone-9B-GGUF: Same weights, converted
- DevQuasar/zai-org.AutoGLM-Phone-9B-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.