V-ICAL Bench: leaderboard
Metric: Normalized score (0-100; environment-weighted). Source: arxiv.org. Saturation forecast: Around October 2027. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Seed 2.1 Pro | 54.4 |
| 2 | Gemini 3.6 Flash | 49.5 |
| 3 | Gemini 3.1 Pro (Preview) | 48 |
| 4 | GPT-5.6 Sol | 46.7 |
| 5 | Gemini 3 Flash | 39.6 |
| 6 | Seed 2.0 Lite | 37 |
| 7 | DeepSeek V4.1 Flash | 34.8 |
| 8 | Qwen 3.5 397B A17B | 32.5 |
| 9 | Kimi K3 | 32.5 |
| 10 | Gemma 4 31B (IT) | 21.9 |
| 11 | MiMo-V2.5 | 20.4 |
| 12 | Qwen 2.5 VL 72B Instruct | 17.7 |
| 13 | Qwen 3.6 27B | 16.9 |
| 14 | GLM-4.5V | 16.6 |
| 15 | Qwen 3.5 27B | 16.4 |
Interactive version: theaggregate.ai/benchmark?slug=v-ical-bench · How It Works · Data refreshed daily, snapshot 2026-09-25.