mradermacher/Qwen3-VL-8B-Thinking-heretic-GGUF

Same weights, converted (fewer layers) vs sh0ck0r/Qwen3-VL-8B-Thinking-heretic. mradermacher/Qwen3-VL-8B-Thinking-heretic-GGUF matches part of sh0ck0r/Qwen3-VL-8B-Thinking-heretic (399 of its 750 tensors; the rest of sh0ck0r/Qwen3-VL-8B-Thinking-heretic 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

On the Hub · Downloads on Model Pulse