mradermacher/PAPO-G-Qwen3-VL-2B-Thinking-GGUF

Same weights, converted (fewer layers) vs PAPOGalaxy/PAPO-G-Qwen3-VL-2B-Thinking. mradermacher/PAPO-G-Qwen3-VL-2B-Thinking-GGUF matches part of PAPOGalaxy/PAPO-G-Qwen3-VL-2B-Thinking (311 of its 626 tensors; the rest of PAPOGalaxy/PAPO-G-Qwen3-VL-2B-Thinking has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/113, MLP 0/84, Attention 0/112

On the Hub · Downloads on Model Pulse