mradermacher/Nanonets-OCR-s-Fast-Preprocessor-GGUF
Same weights, converted (fewer layers) vs amazingvince/Nanonets-OCR-s-Fast-Preprocessor. mradermacher/Nanonets-OCR-s-Fast-Preprocessor-GGUF matches part of amazingvince/Nanonets-OCR-s-Fast-Preprocessor (434 of its 824 tensors; the rest of amazingvince/Nanonets-OCR-s-Fast-Preprocessor has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).
Tensors changed: Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 0/252