whitecircle/GLM-4.7-Flash-Coder

Full fine-tune vs zai-org/GLM-4.7-Flash. Fine-tune: 97% of the weights changed, across 47 of 48 layers.

Tensors changed: Embeddings 0/1, Norms 189/189, MLP 2458/2505, Attention 235/235, lm_head 1/1

3 derivatives checked

• ngquocvinh/GLM-4.7-Flash-Coder-GGUF: Quantization
• LakoMoor/GLM-4.7-Flash-Coder-GGUF: Quantization
• ali-sys-1370/GLM-4.7-Flash-Coder-Q4_K_M-GGUF: Quantization

On the Hub · Downloads on Model Pulse