unsloth/GLM-4.7-Flash-REAP-23B-A3B-GGUF

Same weights, converted vs cerebras/GLM-4.7-Flash-REAP-23B-A3B. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 278 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/189, MLP 0/187, Attention 0/188, lm_head 0/1

On the Hub · Downloads on Model Pulse