Cedar - Text-Only - Swahili: leaderboard
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Around March 2028. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-11-20) | 50.43 |
| 2 | Gemini 2.5 Flash (Thinking) | 46.49 |
| 3 | Gemma 3 27B (IT) | 43.11 |
| 4 | Kimi K2 | 41.17 |
| 5 | Claude Sonnet 4.5 | 38.3 |
| 6 | Qwen 3 VL 235B A22B (Thinking) | 38.08 |
| 7 | Qwen 3 VL 8B (Thinking) | 18.24 |
Interactive version: theaggregate.ai/benchmark?slug=cedar-text-only-swahili · How It Works · Data refreshed daily, snapshot 2026-09-25.