SEA-Vision - Document Parsing (Thai): leaderboard

Metric: Normalized edit distance (0-1) between the parsed page and the annotation on the Thai pages of SEA-Vision end-to-end document parsing (15,234 pages of nine document types across 11 Southeast Asian languages, parsed to structured text, tables, formulas and reading order and scored OmniDocBench-style); lower is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Pro0.11#145
2Qwen 2.5 VL 72B Instruct0.11#364
3Qwen 3 VL 32B Instruct0.11#276
4GPT-4o0.22#333

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.

Interactive version: theaggregate.ai/benchmark?slug=sea-vision-document-parsing-thai · How It Works · Data refreshed daily, snapshot 2026-10-11.