Views
No views yet
proxima-ocr-d.markdown-post3.0.l is an experimental document AI multimodal model fine-tuned on top of Qwen3-VL-8B-Instruct, optimized for high precision OCR and structured document reconstruction. The model converts documents into Markdown, HTML-Markdown, and hybrid enriched documentation formats capable of embedding inline programming languages and reconstructing complex layouts such as tables, forms, and mathematical content. making it ideal for technical documentation, knowledge extraction, and agentic workflows on moderate hardware. This post-3.0 large variant offers enhanced accuracy over smaller siblings for demanding tasks, supporting efficient scaling via Transformers or vLLM for privacy-focused, local deployment in automated parsing pipelines.
| File Name | Quant Type | File Size | File Link |
|---|---|---|---|
| Proxima-OCR-D.Markdown-Post3.0.l.BF16.gguf | BF16 | 16.4 GB | Download |
| Proxima-OCR-D.Markdown-Post3.0.l.F16.gguf | F16 | 16.4 GB | Download |
| Proxima-OCR-D.Markdown-Post3.0.l.Q8_0.gguf | Q8_0 | 8.71 GB | Download |
| Proxima-OCR-D.Markdown-Post3.0.l.mmproj-bf16.gguf | mmproj-bf16 | 1.16 GB | Download |
| Proxima-OCR-D.Markdown-Post3.0.l.mmproj-f16.gguf | mmproj-f16 | 1.16 GB | Download |
| Proxima-OCR-D.Markdown-Post3.0.l.mmproj-q8_0.gguf | mmproj-q8_0 | 752 MB | Download |
