mradermacher/Qwen3-VL-4b-Heretic-GGUF

Same weights, converted (fewer layers) vs DreamFast/Qwen3-VL-4b-Heretic. mradermacher/Qwen3-VL-4b-Heretic-GGUF matches part of DreamFast/Qwen3-VL-4b-Heretic (398 of its 713 tensors; the rest of DreamFast/Qwen3-VL-4b-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

On the Hub · Downloads on Model Pulse