mradermacher/ocr-qwen-GGUF
Same weights, converted (fewer layers) vs syntheticbot/ocr-qwen. mradermacher/ocr-qwen-GGUF matches part of syntheticbot/ocr-qwen (339 of its 729 tensors; the rest of syntheticbot/ocr-qwen has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 0/196, lm_head 0/1