Youtu-Parsing Models by Tencent AI Lab
Tencent AI Lab releases · 5 in the last 12 months · this family litChangelog →
1 model2026
Last refreshed 2026-10-09. Next refresh: weekly.
Details
ResearcherTencent AI Lab
LicenseYoutu-Parsing License
Commercial useCommercial use: conditional
Models1
Released2026
Capabilities
VisionAll models
MultimodalAll models
Structured OutputsAll models
About
Youtu-Parsing is Tencent Youtu Lab's family of open-weight parsing models that turn documents and other media into structured output. Youtu-Parsing-Omni (October 2026) is a 5B omni-modal model that reads document pages, natural images, charts, flowcharts, geometry figures, audio and video and returns one structured JSON format covering layout, text, tables, formulas, timestamps, speech and captions.
Decision facts
- Best fit
- vision and multimodal workstructured outputscoding
- Capability starting point
- Youtu-Parsing-Omni with structured outputs and multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- Hunyuan Image
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
Youtu-Parsing-OmniCurrent
Use when the workload needs multimodal, 5B parameters, and structured outputs.
2026-10multimodal5B parametersstructured outputs
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Youtu-Parsing-Omni | Use when the workload needs multimodal, 5B parameters, and structured outputs. | 2026-10 | multimodal5B parametersstructured outputs | Current |
Release Timeline
1 release group2026-10
1 current
Youtu-Parsing-Omni
Currentmultimodal5B parametersstructured outputs
Specifications(1 models)
| Model | Released | Parameters | Vision | Multimodal | Structured Outputs |
|---|---|---|---|---|---|
| Youtu-Parsing-Omni | 2026-10 | 5B | Yes | Yes | Yes |
