n0kovo/Qwen3-VL-32B-Thinking-heretic-v2-GGUF

Same weights, converted (fewer layers) vs coder3101/Qwen3-VL-32B-Thinking-heretic-v2. n0kovo/Qwen3-VL-32B-Thinking-heretic-v2-GGUF matches part of coder3101/Qwen3-VL-32B-Thinking-heretic-v2 (707 of its 1058 tensors; the rest of coder3101/Qwen3-VL-32B-Thinking-heretic-v2 has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/257, MLP 0/192, Attention 0/256, lm_head 0/1

On the Hub · Downloads on Model Pulse