CorpusQA (128K) - Financial (Chinese): leaderboard
Chinese financial-report domain of CorpusQA (Lu et al., Tongyi Lab, 2026) at the 128K-token scale (concatenated corpora of about 92K tokens): 83 analytical questions (filters, rankings, counts and computed values) answered over the whole concatenated corpus of Chinese financial reports, with reference answers computed by SQL over fields extracted from every document. Accuracy in percent as judged by DeepSeek-V3 against the reference; higher is better.
Metric: Accuracy (%; 83 questions over Chinese financial reports, 128K-token corpora, DeepSeek-V3 judge). Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 93.98 | #163 |
| 2 | GPT-5 Mini (2025-08-07) | 93.98 | #341 |
| 3 | Qwen 3 235B A22B 2507 (Thinking) | 90.36 | #460 |
| 4 | Gemini 2.5 Pro | 89.16 | #262 |
| 5 | Gemini 2.5 Flash | 83.13 | #452 |
| 6 | DeepSeek R1 0528 | 83.13 | #425 |
| 7 | Qwen 3 30B A3B 2507 (Thinking) | 83.13 | #759 |
| 8 | MiniMax M1 80k | 73.49 | #926 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=corpusqa-128k-financial-chinese · How It Works · Data refreshed daily, snapshot 2026-11-12.