π§
OvisOCR2 model by Ath Maas
β 59.5
π¬Technical Deep Dive
Full Specifications [+]
π Updated daily
Source summary: Based on Hugging Face metadata. Not a recommendation.
π‘οΈ Model Transparency Report
Technical metadata sourced from upstream repositories.
Open Metadata
π Identity & Source
- id
- hf-model--ath-maas--ovisocr2
- slug
- ath-maas--ovisocr2
- source
- huggingface
- author
- Ath Maas
- license
- Apache-2.0
- tags
- transformers, safetensors, qwen3_5, image-text-to-text, ocr, document-parsing, multimodal, markdown, tables, formulas, vllm, conversational, arxiv:2607.13639, base_model:qwen/qwen3.5-0.8b, base_model:finetune:qwen/qwen3.5-0.8b, license:apache-2.0, endpoints_compatible, region:us, eval-results
βοΈ Technical Specs
- architecture
- Qwen3_5ForConditionalGeneration
- params billions
- 0.85
- context length
- 32,768
- pipeline tag
- image-text-to-text
- vram gb
- 3.1
- vram is estimated
- true
- vram formula
- VRAM β (params * 0.75) + 2GB (KV) + 0.5GB (OS)
π Engagement & Metrics
- downloads
- 112,513
- stars
- null
- forks
- null
Data indexed from public sources. Updated daily.