OpenAI's GPT-6 Astra, launched on September 3, scored 97.6% on FrontierMath Tier 4 (v2), up from an internal version's 17% and predecessor GPT-5.6 Sol's 83.0%. The model also achieved 99.9% on ARC-AGI…
OpenAI announced GPT-6 Astra, which it claims is its most capable and aligned model, achieving 98% on FrontierMath Tier 4, 99.9% on ARC-AGI-3, and 100% on ExploitBench, and can autonomously control de…
OpenAI launched GPT-6 Astra, which the company says outperforms Anthropic's Fable 5.1 across benchmarks, saturating tests like FrontierMath and ARC-AGI 3, and OpenAI president Greg Brockman hailed it …
OpenAI's GPT-6 Astra, announced on September 3, 2026, reports a 72.6% score on the OSWorld 2.0 benchmark, with tasks taking roughly 40 minutes each, highlighting the shift from short agent runs to lon…
OpenAI launched GPT-6 Astra, a supermodel that beats Fable 5.1 on benchmarks including FrontierMath (97.6%) and ARC-AGI-3 (99.9%), and is capable of acting as an automated AI engineer for under $6 an …
OpenAI launched GPT-6 Astra, which it calls 'the world's most intelligent and aligned model,' with state-of-the-art performance on computer use, browsing, software engineering, cybersecurity, science,…
OpenAI's GPT-6 Astra, rolling out Thursday, outperforms Anthropic's Claude Fable 5.1 on most leaked benchmarks, including a 97.6% score on FrontierMath Tier 4 (v2) versus Fable 5.1's 87.8%, with OpenA…
OpenAI has announced GPT-6 Astra, a major upgrade to ChatGPT and Codex, calling it 'the world's most intelligent and aligned model.' The model scores 98% on FrontierMath Tier 4, 99.9% on ARC-AGI-3, an…
Epoch AI's FrontierMath benchmark now includes roughly 50 unsolved research-level mathematics problems in its Open Problems track, and AI systems have solved exactly three of them, a 6% success rate. …
AGI Ranker, an independent AI leaderboard, released v2.0.0 of its AGI Score, causing every model's score to drop by 6 to 15 points due to a correction in how human-parity ceilings are applied. The rev…
Anthropic's Claude Fable 5 and OpenAI's GPT-5.5 Pro each generated verified solutions to an unsolved number theory problem involving the absolute Galois group of the 2-adic numbers, marking the second…
A third of official solutions in Epoch's FrontierMath benchmark contained errors, making reward-hacking the only way to pass, according to a post by LessWrong user 1a3orn. The author argues that flawe…
Anthropic's Claude Fable 5 scored 161 on the Epoch Capabilities Index, surpassing OpenAI's GPT-5.5 Pro at 159, with dominant performances in math and software engineering benchmarks. The model achieve…
Anthropic's Claude Fable 5 achieved 88 percent accuracy on the hardest FrontierMath tier, surpassing OpenAI's GPT-5.5 by 13 points. The result marks a dramatic improvement from Anthropic's previous Op…
Epoch AI disclosed on May 11, 2026, that an internal audit of its FrontierMath benchmark, a 350-problem test developed with over 60 mathematicians to evaluate AI reasoning, found fatal errors in rough…
Cognition introduced FrontierCode, a new benchmark that evaluates code on mergeability rather than just unit-test passing, with tasks built by open-source maintainers requiring over 40 hours each. The…