FINESSE-Bench - VLigaBench-ru: leaderboard
Metric: Accuracy (%) on the 324 VLigaBench-ru Russian-language olympiad problems in micro- and macroeconomics, financial mathematics and game theory (numerical and short answers) of FINESSE-Bench; a GPT-5.2 judge marks each answer correct or incorrect against the reference; zero-shot, one fixed prompt per task type, temperature 0 where possible, reasoning configurations with medium effort where the model offers them; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 397B A17B | 84.88 |
| 2 | GPT-5.2 (Medium) | 84.88 |
| 3 | Qwen 3.5 Plus (2026-02-15) | 84.88 |
| 4 | Qwen 3.5 122B A10B | 83.64 |
| 5 | Kimi K2.5 | 83.02 |
| 6 | GLM-4.7 | 82.41 |
| 7 | GLM-5 | 82.1 |
| 8 | Qwen 3.5 27B | 81.17 |
| 9 | Qwen 3.5 Flash (02-23) | 81.17 |
| 10 | Qwen 3.5 35B A3B | 80.56 |
| 11 | Claude Sonnet 4.6 (Medium) | 80.56 |
| 12 | Qwen 3 235B A22B 2507 (Thinking) | 79.32 |
| 13 | GPT-5.4 (Medium) | 79.32 |
| 14 | MiniMax-M2.5 | 79.01 |
| 15 | Claude 3.7 Sonnet (Thinking) | 79.01 |
Interactive version: theaggregate.ai/benchmark?slug=finesse-bench-vligabench-ru · How It Works · Data refreshed daily, snapshot 2026-10-07.