mradermacher/DeepPrep-Qwen3-8B-GGUF

Same weights, converted vs DeepPrep/DeepPrep-Qwen3-8B. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/145, Attention 0/144, MLP 0/108

On the Hub · Downloads on Model Pulse