mradermacher/Dots.OCR-Latest-BF16-GGUF
Same weights, converted (fewer layers) vs prithivMLmods/Dots.OCR-Latest-BF16. mradermacher/Dots.OCR-Latest-BF16-GGUF matches part of prithivMLmods/Dots.OCR-Latest-BF16 (339 of its 643 tensors; the rest of prithivMLmods/Dots.OCR-Latest-BF16 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