The ARC Prize Foundation is a nonprofit organization dedicated to open scientific progress through enduring AI benchmarks.
We make tools that provide empirical data about intelligence capabilities which guide critical industry decisions about research, safety, and policy.
The ARC Prize benchmarks are designed to measure AI progress, not to serve as a target for optimization. ARC Prize tasks are not economically useful to target, instead they are a measure of AI capability. Our goal is to provide an objective assessment of model capabilities rather than influence model training through repeated, iterative testing.
We do this in three ways:
Providing open source datasets (ARC-AGI-1 and ARC-AGI-2) and software for benchmarking model capabilties.
Keeping an up-to-date leaderboard of state-of-the-art model performance on "semi-private" ARC-AGI task datasets. This leaderboard has no limitations on internet access or compute, and is intended to test state of the art models and bespoke solutions.
Hosting an annual competition on an additional "private" set of ARC-AGI tasks for open-source models without bounded compute and no internet access.
We test new models that are publicly released as well as proprietary models we're given open access to.
All models we test appear on our leaderboard, and all models on the leaderboard have been tested by us on the semi-private test set. Only aggregate scores are shared publicly. Average cost per task is calculated and reported. Model providers are notified when possible.
Additionally, we test publicly available models on the public evaluation task set. Full results including model outputs, evaluation durations, costs, and individual task scores are shared alongside the overall model score on HuggingFace..
Note: The ARC-AGI Private Test Set is reserved exclusively for use in the annual competition to prevent data leakage.
We collaborate with major open-source and commercial model providers to test unreleased models for the community. A model is considered unreleased if its weights are neither open nor available via a public API or service. The models that are tested must be intended for public launch.
Our approach to testing unreleased models:
This approach ensures that ARC Prize remains an independent evaluator of frontier AI capabilities.
Many researchers develop custom solutions to ARC-AGI. Examples include submissions from MIT/Cornell and Jeremy Berman.
Bespoke Solution Rules
Submissions
The ARC Prize team will verify solutions and scores within 2 weeks of submission.
Only new high score claims (e.g., submissions that exceed the current high score by at least 1%) will be verified and published. This is to limit the semi-private evaluation set from leaking into LLM training data.
If you have a solution that is expensive to run, we encourage you to test by randomly sampling 100 tasks from the 400 public evaluation tasks and hold out the remainder for private validation. This can build confidence towards your overall ARC-AGI-Pub score before incurring significant cost with the full 400 task dataset.
Submit an ARC-AGI-Pub high score claim.
Compute and/or provider costs can be significant to run solutions against evaluation sets. To help support those contributing to this initiative, we’ve set up a verification fund.
ARC-AGI submission to be verified can use up to $10,000 in resources over the 500 puzzles (400 public evaluation set + 100 semi-private evaluation set.). We retain the right to refuse verification of any previous benchmarks in the event of future benchmark releases (such as ARC-AGI-2 or 3).
For each new proven SOTA reproduction (i.e., when a solution scores higher than the current highest score on this leaderboard on the public evaluation set), we will reimburse up to $2,500.
This fund is a work-in-progress and we reserve the right to make changes at any time or refuse reimbursement requests upon consideration by the ARC Prize team.
We will continuously add new models and retire old ones. It is not feasible to add every possible model due to the cost and the scalability of our evaluation process.
We want to assess performance across different levels of reasoning. To do this, we will repeat model tests at spaced reasoning levels.
No it doesn't. This is a common misconception. The ARC-AGI-Pub leaderboard is open to all models. Here's a example of a prompt that was used to test on ARC-AGI: Prompt
If a model is open source and not available via API by the model creator we will use another public model provider. This includes companies like Baseten or TogetherAI.
Cost is a critical factor in model evaluation, and whenever possible, we will use retail pricing to assess cost efficiency. For model providers, we will base cost calculations on publicly available retail rates—typically measured in price per million tokens—rather than a provider’s internal margins or raw cost of goods. Costs are generally shared on an average per-test-pair-attempt basis.
We are a nonprofit that seeks to provide transparency and all tools. We invite the community to reproduce our results.
The ARC Prize Foundation is funded exclusively through donations, including financial contributions, cloud credits, and API credits.
We do not accept funding from AI model providers for core benchmark and leaderboard evaluations. Our evaluations are conducted independently, and we take potential conflicts of interest very seriously. Our commitment is to scientific rigor, transparency, and impartiality in AI benchmarking.
If you’d like to support our work, please visit our donation page.
Feel free to contact us at: team@arcprize.org