mradermacher/qwen3vl8b-finetune-GGUF
Same weights, converted (fewer layers) vs om-ai/qwen3vl8b-finetune. mradermacher/qwen3vl8b-finetune-GGUF matches part of om-ai/qwen3vl8b-finetune (399 of its 750 tensors; the rest of om-ai/qwen3vl8b-finetune has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/108, Attention 0/144, lm_head 0/1