FCMBench-Video (Chinese) - Evidence-Grounded Selection: leaderboard

FCMBench-Video (2026) Chinese-language handheld document videos from financial credit workflows, task evidence-grounded selection: exact-match accuracy (%) on single-choice questions answered from the right document in a multi-document video. Zero-shot; higher is better.

Metric: Evidence-Grounded Selection: exact-match accuracy (%) on single-choice questions answered from the right document in a multi-document video, on FCMBench-Video's zh-CN subset (405 composed videos of 20 Chinese financial-credit document types, 5,960 instances); zero-shot with structured outputs; each model's native video serving path (2 FPS sampling for most open models, 0.5 FPS for Ovis2.5, raw video upload for the API models); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro (Preview)73.04#119
2Qwen 3.5 27B67.32#375
3Doubao-Seed-1.6-vision50.18#361
4Qwen 3 VL 32B Instruct44.11#606
5Qwen 3 VL 8B Instruct35#1029
6Qwen3 Omni 30B A3B Instruct33.75#930
7InternVL3-8B15.18#1782
8Ovis2.5-9B13.04#1225

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

Interactive version: theaggregate.ai/benchmark?slug=fcmbench-video-chinese-evidence-grounded-selection · How It Works · Data refreshed daily, snapshot 2026-11-21.