mradermacher/KarantaOCR-GGUF

Same weights, converted (fewer layers) vs taresco/KarantaOCR. mradermacher/KarantaOCR-GGUF matches part of taresco/KarantaOCR (434 of its 824 tensors; the rest of taresco/KarantaOCR has no counterpart). On the shared part: Same weights stored in another dtype (F32 → F16/F32).

Tensors changed: Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 0/252

On the Hub · Downloads on Model Pulse