AI Leaderboards
Explore AI rankings by metric.
Data as of 11/3/2026 · Current update pending; showing last verified data
5 of 7 leaderboards with verified data
Leaderboard
See who leads under one clear measure
TOP 10 · Current profile
Context capacity
Maximum context length in the official model config
- Unit
- tokens
- Coverage
- 18 objects
- Updated
- Excluded
- 42 records
Scroll sideways to review all leaderboard fields.
| Rank | Model | Maximum context length in the official model config | Data date |
|---|---|---|---|
| 1 | DeepSeek R1 | 163,840tokens | |
| 1 | DeepSeek V3 | 163,840tokens | |
| 3 | DeepSeek R1 Distill Qwen 32B | 131,072tokens | |
| 3 | GPT OSS 120B on Groq | 131,072tokens | |
| 3 | Mistral Nemo Instruct 2407 | 131,072tokens | |
| 6 | Qwen2.5 72B Instruct | 32,768tokens | |
| 6 | Qwen2.5 7B Instruct | 32,768tokens | |
| 6 | Qwen2.5 Coder 32B Instruct | 32,768tokens | |
| 6 | Zephyr 7B Beta | 32,768tokens | |
| 6 | Mistral 7B v0.1 | 32,768tokens |