mradermacher/PAPO-D-Qwen2.5-VL-7B-GGUF

Same weights, converted (fewer layers) vs PAPO-Galaxy/PAPO-D-Qwen2.5-VL-7B. mradermacher/PAPO-D-Qwen2.5-VL-7B-GGUF matches part of PAPO-Galaxy/PAPO-D-Qwen2.5-VL-7B (339 of its 729 tensors; the rest of PAPO-Galaxy/PAPO-D-Qwen2.5-VL-7B has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Norms 0/57, MLP 0/84, Attention 0/196, Embeddings 0/1, lm_head 0/1

On the Hub · Downloads on Model Pulse