Anthropic has released Claude Fable 5.2 and Claude Mythos 5.2, with Fable 5.2 now available to all developers via API and cloud platforms. Both models feature a 1M-token context window and improved speed, but Mythos 5.2 is restricted to verified organizations due to its advanced capabilities in security and biology research. Anthropic highlighted Fable 5.2's ability to coordinate complex, long-running tasks, citing a demonstration where 1,000 agents formalized Fermat's Last Theorem in 30 hours, significantly outperforming Opus 5.5 in a similar setup. AI
IMPACT Sets new SOTA on long-running agent tasks and coding benchmarks, potentially accelerating enterprise adoption of complex AI workflows.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
- Amazon Bedrock
- Anthropic
- Claude API
- Claude Fable 5.1
- CNBC
- Fermat's Last Theorem
- Google Cloud
- GPT-6 Astra
- Lean
- Microsoft Foundry
- Opus 5.5
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →