SuperCLUE-Ondevice (September 2025) - Language: leaderboard
SuperCLUE-OnDevice evaluates on-device phone-class models; the roster also scores several large cloud models as reference points. September 2025 edition (2025年9月): language sheet total covering summarisation, text creation, information extraction, instruction following and hallucination control. Each dated workbook is a separate evaluation round, kept apart from the current board. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 82.77 | #159 |
| 2 | Doubao-Seed-1.6-thinking-250715 | 80.32 | #169 |
| 3 | GLM-4.5V | 76.45 | #358 |
| 4 | Qwen 2.5 VL 7B | 67.82 | #641 |
| 5 | Gemma 3 4B | 66.9 | #1088 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-ondevice-september-2025-language · How It Works · Data refreshed daily, snapshot 2026-11-05.