[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.] [@arcprize](/creator/twitter/arcprize) "o3 Pro on ARC-AGI Semi Private Eval Results ARC-AGI-1: * Low: XX% $1.64/task * Medium: XX% $3.18/task * High: XX% $4.16/task ARC-AGI-2: * All reasoning efforts: X% $4-7/task Takeaways: * o3-pro in line with o3 performance * o3's new price sets the ARC-AGI-1 Frontier"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1932535378080395332) 2025-06-10 20:28:33 UTC 24.6K followers, 125.8K engagements "We're excited to share this preview of ARC-AGI-3. This is just the beginning Visit to learn more"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260385398722761) 2025-07-18 17:26:50 UTC 24.6K followers, 7761 engagements "ARC-AGI-3 Developer Preview * Hands on first look at ARC-AGI-3 (live demos & API access) * Fireside with @fchollet moderated by @dwarkesh_sp 7/17 San Francisco Open to sponsors & researchers of @arcprize (very limited public slots available)"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1939723351335047515) 2025-06-30 16:30:59 UTC 24.6K followers, 52.7K engagements "New ARC Prize 2025 High Score XXXX% by @MindsAI_Jack @MohamedOsmanML @tufalabs"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1942247297301275058) 2025-07-07 15:40:15 UTC 24.6K followers, 29.6K engagements "We create benchmarks that highlight the gap between human generalization and machine patternmatching ARCAGI1 (2019) challenged deeplearning ARCAGI2 (2025) challenges static reasoning models"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260368332104148) 2025-07-18 17:26:46 UTC 24.6K followers, 8881 engagements "Today we're announcing a preview of ARC-AGI-3 the Interactive Reasoning Benchmark with the widest gap between easy for humans and hard for AI Were releasing: * X games (environments) * $10K agent contest * AI agents API Starting scores - Frontier AI: X% Humans: 100%"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260363256996244) 2025-07-18 17:26:45 UTC 24.6K followers, 282.7K engagements "Reported scores are from ARC-AGI-1 & X Semi Private Evaluation Set Learn more about ARC Prize Foundation: ARC-AGI Reasoning Frontier Blog Post: View the leaderboard: Reproduce the results:"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1943168960280826274) 2025-07-10 04:42:37 UTC 24.6K followers, 44.8K engagements ".@fchollet presenting at @ycombinator Start Up School He presents on: * The path to AGI * ARC-AGI-3 * How intelligence is building *new skills* not maximizing memorization"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1940794238519930919) 2025-07-03 15:26:19 UTC 24.6K followers, 21.6K engagements "Hear @GregKamradt talk about ARC-AGI-3 with @swyx and @FanaHOVA on @latentspacepod * Why interactive benchmarks * Defining Intelligence * Play through of ARC-AGI-3 games"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946311135990608168) 2025-07-18 20:48:30 UTC 24.6K followers, 7410 engagements "Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI by @PourcelJulien @cedcolas and @pyoudeyer Another example of ARC-AGI as a research playground that has general applicability"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1944814215665213767) 2025-07-14 17:40:16 UTC 24.6K followers, 9879 engagements "Grok X (Thinking) achieves new SOTA on ARC-AGI-2 with XXXX% This nearly doubles the previous commercial SOTA and tops the current Kaggle competition SOTA"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1943168950763950555) 2025-07-10 04:42:34 UTC 24.6K followers, 7.2M engagements "Agents are now the frontier. They perceive plan act remember adapt. Static puzzles arent equipped to grade that loop We need interactive benchmarks that test worldmodel building and longhorizon planning under sparse feedback"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260371289088444) 2025-07-18 17:26:46 UTC 24.6K followers, 7673 engagements "Enter ARCAGI3 X brandnew games. Easy for humans out of reach for todays best AI models X games are live today X will go live in August"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260374590005308) 2025-07-18 17:26:47 UTC 24.6K followers, 7219 engagements "Every game environment is novel unique and only requires core-knowledge priors No language trivia or specialized knowledge is needed to beat the games * Play: * Compete: * Build:"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260365110833320) 2025-07-18 17:26:45 UTC 24.6K followers, 10.9K engagements "Interactive Reasoning Benchmarks are the next step in frontier evaluations Hear @GregKamradt share why measuring human-like intelligence requires multi-turn environments Including a sneak peak of ARC-AGI-3 Want to help us build interactive evaluations We're hiring"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1932137879742063073) 2025-06-09 18:09:02 UTC 24.6K followers, 27.2K engagements "ARC-AGI-3 Preview games need to be pressure tested. Were hosting a 30-day agent competition in partnership with @huggingface Were calling on the community to build agents (and win money)"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260382890463487) 2025-07-18 17:26:49 UTC 24.6K followers, 21K engagements "On ARC-AGI-1 Grok X (Thinking) achieves XXXX% inline with the Pareto frontier for AI reasoning systems we reported last month"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1943168952936591809) 2025-07-10 04:42:35 UTC 24.6K followers, 63.7K engagements "Gemini XXX Pro (6/17) on ARC-AGI Semi Private Eval ARC-AGI-1: * Thinking 1K: XX% $0.06/task * Thinking 8K: XX% $0.29/task * Thinking 16K: XX% $0.48/task * Thinking 32K: XX% $0.51/task ARC-AGI-2: * Thinking 32K: XXX% $0.75/task"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1936140215556751415) 2025-06-20 19:12:53 UTC 24.6K followers, 52.3K engagements "Your ability to efficiently adapt to novelty defines your intelligence not your performance on a single-skill Harder puzzles dont prove smarter AI but rather its ability to learn new rules does ARC Prize exists to operationalize that insight"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260366838931919) 2025-07-18 17:26:45 UTC 24.6K followers, 10.1K engagements "ARC-AGI API ships today Plug in any LLM RL or hybrid agent train locally test against our servers"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260376951398590) 2025-07-18 17:26:48 UTC 24.6K followers, 7762 engagements "Claude Opus X on ARC-AGI Semi Private Eval Base * ARC-AGI-1: XXXX% $0.40/task * ARC-AGI-2: XXX% $0.63/task Thinking 16K * ARC-AGI-1: XXXX% $1.25/task * ARC-AGI-2: XXX% $1.93/task Opus X sets new SOTA (8.6%) on ARC-AGI-2"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1927817468342304848) 2025-05-28 20:01:16 UTC 24.6K followers, 55.3K engagements "Thank you to the @xai team for working with us to validate Grok 4's score and inviting us to the watch the live stream"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1943168955956572649) 2025-07-10 04:42:36 UTC 24.6K followers, 78.6K engagements "o3 (left) and Grok X (right) replays below spoiler: neither complete a single level"  [@arcprize](/creator/x/arcprize) on [X](/post/tweet/1946260379405066372) 2025-07-18 17:26:48 UTC 24.6K followers, 71.3K engagements
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]
@arcprize
"o3 Pro on ARC-AGI Semi Private Eval Results ARC-AGI-1: * Low: XX% $1.64/task * Medium: XX% $3.18/task * High: XX% $4.16/task ARC-AGI-2: * All reasoning efforts: X% $4-7/task Takeaways: * o3-pro in line with o3 performance * o3's new price sets the ARC-AGI-1 Frontier" @arcprize on X 2025-06-10 20:28:33 UTC 24.6K followers, 125.8K engagements
"We're excited to share this preview of ARC-AGI-3. This is just the beginning Visit to learn more" @arcprize on X 2025-07-18 17:26:50 UTC 24.6K followers, 7761 engagements
"ARC-AGI-3 Developer Preview * Hands on first look at ARC-AGI-3 (live demos & API access) * Fireside with @fchollet moderated by @dwarkesh_sp 7/17 San Francisco Open to sponsors & researchers of @arcprize (very limited public slots available)" @arcprize on X 2025-06-30 16:30:59 UTC 24.6K followers, 52.7K engagements
"New ARC Prize 2025 High Score XXXX% by @MindsAI_Jack @MohamedOsmanML @tufalabs" @arcprize on X 2025-07-07 15:40:15 UTC 24.6K followers, 29.6K engagements
"We create benchmarks that highlight the gap between human generalization and machine patternmatching ARCAGI1 (2019) challenged deeplearning ARCAGI2 (2025) challenges static reasoning models" @arcprize on X 2025-07-18 17:26:46 UTC 24.6K followers, 8881 engagements
"Today we're announcing a preview of ARC-AGI-3 the Interactive Reasoning Benchmark with the widest gap between easy for humans and hard for AI Were releasing: * X games (environments) * $10K agent contest * AI agents API Starting scores - Frontier AI: X% Humans: 100%" @arcprize on X 2025-07-18 17:26:45 UTC 24.6K followers, 282.7K engagements
"Reported scores are from ARC-AGI-1 & X Semi Private Evaluation Set Learn more about ARC Prize Foundation: ARC-AGI Reasoning Frontier Blog Post: View the leaderboard: Reproduce the results:" @arcprize on X 2025-07-10 04:42:37 UTC 24.6K followers, 44.8K engagements
".@fchollet presenting at @ycombinator Start Up School He presents on: * The path to AGI * ARC-AGI-3 * How intelligence is building new skills not maximizing memorization" @arcprize on X 2025-07-03 15:26:19 UTC 24.6K followers, 21.6K engagements
"Hear @GregKamradt talk about ARC-AGI-3 with @swyx and @FanaHOVA on @latentspacepod * Why interactive benchmarks * Defining Intelligence * Play through of ARC-AGI-3 games" @arcprize on X 2025-07-18 20:48:30 UTC 24.6K followers, 7410 engagements
"Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI by @PourcelJulien @cedcolas and @pyoudeyer Another example of ARC-AGI as a research playground that has general applicability" @arcprize on X 2025-07-14 17:40:16 UTC 24.6K followers, 9879 engagements
"Grok X (Thinking) achieves new SOTA on ARC-AGI-2 with XXXX% This nearly doubles the previous commercial SOTA and tops the current Kaggle competition SOTA" @arcprize on X 2025-07-10 04:42:34 UTC 24.6K followers, 7.2M engagements
"Agents are now the frontier. They perceive plan act remember adapt. Static puzzles arent equipped to grade that loop We need interactive benchmarks that test worldmodel building and longhorizon planning under sparse feedback" @arcprize on X 2025-07-18 17:26:46 UTC 24.6K followers, 7673 engagements
"Enter ARCAGI3 X brandnew games. Easy for humans out of reach for todays best AI models X games are live today X will go live in August" @arcprize on X 2025-07-18 17:26:47 UTC 24.6K followers, 7219 engagements
"Every game environment is novel unique and only requires core-knowledge priors No language trivia or specialized knowledge is needed to beat the games * Play: * Compete: * Build:" @arcprize on X 2025-07-18 17:26:45 UTC 24.6K followers, 10.9K engagements
"Interactive Reasoning Benchmarks are the next step in frontier evaluations Hear @GregKamradt share why measuring human-like intelligence requires multi-turn environments Including a sneak peak of ARC-AGI-3 Want to help us build interactive evaluations We're hiring" @arcprize on X 2025-06-09 18:09:02 UTC 24.6K followers, 27.2K engagements
"ARC-AGI-3 Preview games need to be pressure tested. Were hosting a 30-day agent competition in partnership with @huggingface Were calling on the community to build agents (and win money)" @arcprize on X 2025-07-18 17:26:49 UTC 24.6K followers, 21K engagements
"On ARC-AGI-1 Grok X (Thinking) achieves XXXX% inline with the Pareto frontier for AI reasoning systems we reported last month" @arcprize on X 2025-07-10 04:42:35 UTC 24.6K followers, 63.7K engagements
"Gemini XXX Pro (6/17) on ARC-AGI Semi Private Eval ARC-AGI-1: * Thinking 1K: XX% $0.06/task * Thinking 8K: XX% $0.29/task * Thinking 16K: XX% $0.48/task * Thinking 32K: XX% $0.51/task ARC-AGI-2: * Thinking 32K: XXX% $0.75/task" @arcprize on X 2025-06-20 19:12:53 UTC 24.6K followers, 52.3K engagements
"Your ability to efficiently adapt to novelty defines your intelligence not your performance on a single-skill Harder puzzles dont prove smarter AI but rather its ability to learn new rules does ARC Prize exists to operationalize that insight" @arcprize on X 2025-07-18 17:26:45 UTC 24.6K followers, 10.1K engagements
"ARC-AGI API ships today Plug in any LLM RL or hybrid agent train locally test against our servers" @arcprize on X 2025-07-18 17:26:48 UTC 24.6K followers, 7762 engagements
"Claude Opus X on ARC-AGI Semi Private Eval Base * ARC-AGI-1: XXXX% $0.40/task * ARC-AGI-2: XXX% $0.63/task Thinking 16K * ARC-AGI-1: XXXX% $1.25/task * ARC-AGI-2: XXX% $1.93/task Opus X sets new SOTA (8.6%) on ARC-AGI-2" @arcprize on X 2025-05-28 20:01:16 UTC 24.6K followers, 55.3K engagements
"Thank you to the @xai team for working with us to validate Grok 4's score and inviting us to the watch the live stream" @arcprize on X 2025-07-10 04:42:36 UTC 24.6K followers, 78.6K engagements
"o3 (left) and Grok X (right) replays below spoiler: neither complete a single level" @arcprize on X 2025-07-18 17:26:48 UTC 24.6K followers, 71.3K engagements
/creator/twitter::1773935160192647168/posts