qinglong-captions v4.7.0: Major Upgrades to Multimodal OCR and Underlying Models
bdsqlsz · x · 2026-08-02
The image captioning tool qinglong-captions has released v4.7.0, bringing several updates to multimodal capabilities and model support:
- OCR & OMR: Introduced OvisOCR2 for image and PDF OCR (supporting Transformers and OpenAI-compatible vLLM backends). Upgraded MuSViT to a complete ONNX sheet-music OMR workflow with MusicXML/MIDI export.
- Model Unification & Updates: Unified Kimi K3 model selection and reasoning controls, and refreshed catalogs for Gemini 3.x and MiniMax M2.7.
- Image Processing: Enforced image prompt-template I/O contracts and synchronized upstream behaviors for LayerDiff and Marigold.
Related event: qinglong-captions v4.7.0 Adds Multimodal OCR and Score Recognition(2 posts)→
More from Multimodal
- Seedance 2.5 workflow: character reference sheets + structured prompt yields a 30s single-take fake tourist video — techhalla · 2026-10-16
- AI couple moves into a castle ruin: full Magnific + Seedance 2.5 prompt thread behind the viral clip — techhalla · 2026-10-16
- FLUX.3 Klein shines at consistent high-res photo editing on RTX 5090: 3000×4000 in ~230s — herbertseabra · 2026-10-16
- Smoky Velvet Noir: a reusable image-gen prompt template with three moody examples — LudovicCreator · 2026-10-16
- VideoGen-Agent: An 8B Multimodal Agent Turns Video Generation Into a Tool-Use Loop — jiqizhixin · 2026-10-16
- Redditor builds impressive seaplane sim project with Sonnet 5.5 + Blender — Wise_Shame_2853 · 2026-10-16