mradermacher/epsilon-ocr-d.markdown-post3.0.m-GGUF

Same weights, converted (fewer layers) vs prithivMLmods/epsilon-ocr-d.markdown-post3.0.m. mradermacher/epsilon-ocr-d.markdown-post3.0.m-GGUF matches part of prithivMLmods/epsilon-ocr-d.markdown-post3.0.m (434 of its 824 tensors; the rest of prithivMLmods/epsilon-ocr-d.markdown-post3.0.m 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

On the Hub · Downloads on Model Pulse