A harness pushes Claude Opus 5 past 30% on ARC-AGI-3

Per the author’s write-up, an agent harness built with Strands took Claude Opus 5 to 99.95% on ARC-AGI-3, against 30% for the model on its own, and that Strands figure isn’t corroborated in indexed sources. The 30% baseline holds up, and NVIDIA reports 100% for Claude Opus 5 wrapped in its own AVO harness.
Up next
HTTPX2 becomes the default in OpenAI’s Python SDK
Certificate verification in the OpenAI Python SDK now leans on the operating-system trust store: HTTPX2 dropped certifi, and minimal container images without system CA certificates can fail outright.
Related stories
- Coding agent clears all 183 ARC-AGI-3 levels
- Claude learned proteins, not just papers about them
- Claude Haiku 5.5 spotted a data leak GPT-6.1 Sol missed
- Fable 5.2 matched Opus 5.5 on nine tasks at almost 4x cost
- Cheaper per token, Fable 5.2 cost $13.39 to Astra’s $3.99
- Managed Agents goes fully self-hosted in private beta
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.

