Benchmark
Orak Supermario
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 6
- Scores recorded
- 10
- Sources
- 1
- IQ weight
- 0.49
Score distribution
- 32.2%
- Mean
- 32.3%
- Median
- 2.9%
- Std dev
- 28.8%
- 10th percentile
- 34.9%
- 90th percentile
- 28.7%
- Min
This benchmark has a weight of 0.49 for chat models, measured over 6 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro Preview 03-25default· Orak leaderboard docs | 38% |
| 2 | o3 Minidefault· Orak leaderboard docs | 34.9% |
| 3 | GPT 4odefault· Orak leaderboard docs | 34.1% |
| 4 | GPT 4oref-plan· Orak leaderboard docs | 34.1% |
| 5 | GPT 4oreflection· Orak leaderboard docs | 32.3% |
| 6 | Claude Sonnet 3.7default· Orak leaderboard docs | 31.7% |
| 7 | GPT 4ozeroshot· Orak leaderboard docs | 29.6% |
| 8 | GPT 4oplanning· Orak leaderboard docs | 29.4% |
| 9 | GPT-4o minidefault· Orak leaderboard docs | 28.8% |
| 10 | DeepSeek R1default· Orak leaderboard docs | 28.7% |