AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
LLM Stats
ARC-AGI v2
ARC-AGI-2 is an upgraded benchmark for measuring abstract reasoning and problem-solving abilities in AI systems through visual grid transformation tasks. It evaluates fluid intelligence via input-output grid pairs (1x1 to 30x30) using colored cells (0-9), requiring models to identify underlying transformation rules from demonstration examples and apply them to test cases. Designed to be easy for humans but challenging for AI, focusing on core cognitive abilities like spatial reasoning, pattern recognition, and compositional generalization.
Evaluation context and evidence
- Source
- LLM Stats
- Publisher
- Not recorded
- Benchmark release
- Not recorded
- Collected
- 2026-08-17
- Modality
- multimodal
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 17
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| GPT-5.5 | 0.85 | Not recorded | Not recorded | 2026-04-23 · model_announcement | LLM Stats |
| Gemini 3.1 Pro | 0.771 | Not recorded | Not recorded | 2026-02-19 · model_announcement | LLM Stats |
| GPT-5.4 | 0.733 | Not recorded | Not recorded | 2026-03-05 · model_announcement | LLM Stats |
| Gemini 3.5 Flash | 0.721 | Not recorded | Not recorded | 2026-05-19 · model_announcement | LLM Stats |
| Claude Opus 4.6 | 0.688 | Not recorded | Not recorded | 2026-02-05 · model_announcement | LLM Stats |
| Claude Sonnet 4.6 | 0.583 | Not recorded | Not recorded | 2026-02-17 · model_announcement | LLM Stats |
| GPT-5.2 Pro | 0.542 | Not recorded | Not recorded | 2025-12-11 · model_announcement | LLM Stats |
| GPT-5.2 | 0.529 | Not recorded | Not recorded | 2025-12-11 · model_announcement | LLM Stats |
| Muse Spark | 0.425 | Not recorded | Not recorded | 2026-04-08 · model_announcement | LLM Stats |
| Inkling-Small | 0.401 | Not recorded | Not recorded | 2026-07-30 · model_announcement | LLM Stats |
| Claude Opus 4.5 | 0.376 | Not recorded | Not recorded | 2025-11-24 · model_announcement | LLM Stats |
| Gemini 3 Flash | 0.336 | Not recorded | Not recorded | 2025-12-17 · model_announcement | LLM Stats |
| Gemini 3 Pro | 0.311 | Not recorded | Not recorded | 2025-11-18 · model_announcement | LLM Stats |
| Grok-4 | 0.159 | Not recorded | Not recorded | 2025-07-09 · model_announcement | LLM Stats |
| Claude Opus 4 | 0.086 | Not recorded | Not recorded | 2025-05-22 · model_announcement | LLM Stats |
| o3 | 0.065 | Not recorded | Not recorded | 2025-04-16 · model_announcement | LLM Stats |
| Gemini 2.5 Pro | 0.049 | Not recorded | Not recorded | 2025-05-20 · model_announcement | LLM Stats |