CFMME - Table Recognition: leaderboard

CFMME (2026) is a Chinese financial multimodal evaluation set for vision-language models, built from financial documents, charts, seals, tables and exam questions. This board is the TEDS tree-edit similarity in percent between recognized and gold tables; higher is better.

Metric: Table recognition TEDS (%, tree-edit-distance similarity of the predicted table structure and content) on CFMME (6,052 Chinese financial multimodal instances over eight financial image types), zero-shot, mean of three runs, the non-reasoning model of each family; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Qwen 3 VL 235B A22B Instruct78.92#532
2GLM-4.5VNon-reasoning71.89#754
3Qwen 3 VL 8B Instruct71.35#1029
4InternVL3.5-241B-A28BNon-reasoning65.6#733
5InternVL3.5-8BNon-reasoning59.96#1604
6ERNIE 4.5 VL 28B A3B57.49#2121
7GPT-4o57.11#751
8Kimi-VL-A3B-Instruct54.2#2253
9Llama 3.2 11B Instruct28.92#5174

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=cfmme-table-recognition · How It Works · Data refreshed daily, snapshot 2026-11-21.