Global-MMLU-Lite leaderboard
As of November 14, 2026, among the results listed here, Thinking Machines Inkling 2 Small has the highest reported Global-MMLU-Lite score, 91.0%, published by Thinking Machines, ahead of Thinking Machines Inkling at 88.7%. 3 models have a Global-MMLU-Lite result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
Best reported score
91.0%
T Thinking Machines Inkling 2 Small
Reported by Thinking Machines, November 9, 2026
3 models ranked · 1 maker
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | T Inkling 2 Small | 91.0% | Not published | Thinking Machines | November 9, 2026 (model release) |
| 2 | T Inkling | 88.7% | Not published | Thinking Machines | July 30, 2026 (model release) |
| 3 | T Inkling Small | 86.7% | Not published | Thinking Machines | July 30, 2026 (model release) |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- All ranked results use the same reported setting.
- Spread
- 91.0% to 86.7% across 3 models.
- Freshness
- Newest result published November 9, 2026; oldest July 30, 2026.
What Global-MMLU-Lite measures
It tests models on culturally sensitive and culture-neutral tasks across languages, with scores indicating performance on this balanced task set. Source: Global MMLU | Multilingual AI Evaluation Benchmark