Humanity's Last Exam
Challenging LLMs at the frontier of human knowledge
25.32±1.70
21.64±1.61
20.32±1.58
Humanity's Last Exam (Text Only)
Challenging LLMs at the frontier of human knowledge
26.32±1.86
22.06±1.75
20.57±1.71
MultiNRC
Multilingual Native Reasoning Evaluation Benchmark for LLMs
52.13±3.01
o3-pro-2025-06-10-high
49.00±3.02
o3-2025-04-16-high
45.50±3.00
MultiChallenge
Assessing models across diverse, interdisciplinary challenges
63.77±1.53
58.55±3.03
59.09±1.08
Fortress
Frontier Risk Evaluation for National Security and Public Safety
8.24±1.93
12.96±2.34
14.79±2.49
MASK
Evaluate model honesty when pressured to lie
Claude Sonnet 4 (Thinking)
95.33±2.29
94.20±1.79
92.00±0.86
EnigmaEval
Evaluating model performance on complex, multi-step reasoning tasks
13.09±1.92
11.91±1.85
10.47±1.74
VISTA
Vision-Language Understanding benchmark for multimodal models
54.65±1.46
54.63±0.55
51.79±0.63
TutorBench
Evaluating model performance on common tutoring tasks for high school and AP-level subjects
gemini-2.5-pro-preview-06-05
55.65±1.11
gpt-5-2025-08-07
55.33±1.02
o3-pro-2025-06-10
54.62±1.02
Frontier AI Model Evaluations & Benchmarks
We conduct high-complexity evaluations to expose model failures, prevent benchmark saturation, and push model capabilities —while continuously evaluating the latest frontier models.
Scaling with Human Expertise
Humans design complex evaluations and define precise criteria to assess models, while LLMs scale evaluations—ensuring efficiency and alignment with human judgment.
Robust Datasets for Reliable AI Benchmarks
Our leaderboards are built on carefully curated evaluation sets, combining private datasets to prevent overfitting and open-source datasets for broad benchmarking and comparability.
Run evaluations on frontier AI capabilities
If you'd like to add your model to this leaderboard or a future version, please contact leaderboards@scale.com. To ensure leaderboard integrity, we require that models can only be featured the FIRST TIME when an organization encounters the prompts.