TabVerse - QA (Markdown Text): leaderboard

TabVerse is a controlled cross-format table-understanding benchmark of 700 balanced question-table pairs drawn from five TableQA sources, with each table aligned across HTML, Markdown and LaTeX as both structured text and rendered images. This board reports Exact-match accuracy (%) answering table questions from Markdown table text; higher is better.

Metric: Exact-match accuracy (%) answering table questions from Markdown table text; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 17 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3 Flash (Preview)65.43#144
2GPT-5.258#176
3Qwen 3 VL 8B Instruct53.29#981
4Gemma 3 27B (IT)53.14#1289
5Gemma 3 12B (IT)48.57#1837
6Qwen 3 30B A3B Instruct46.57#1025
7Qwen 2.5 7B Instruct45.43#2564
8Qwen 3 VL 30B A3B Instruct39.71#871

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=tabverse-qa-markdown-text · How It Works · Data refreshed daily, snapshot 2026-11-13.