mradermacher/DeepSeek-V4-Pro-Qwen3.5-4B-GGUF

Same weights, converted vs Jackrong/DeepSeek-V4-Pro-Qwen3.5-4B. Same weights stored in another dtype (BF16/F32 → F16/F32), plus 39 new tensors.

Tensors changed: Embeddings 0/1, MLP 0/96, Norms 0/81, Attention 0/32

On the Hub · Downloads on Model Pulse