NVIDIA Watch posted an update
Nvidia and ChipAgents have released an evaluation for AI models doing chip-design work, The Information reports. Its top-performing agent, powered by Anthropic’s Fable 5.1, scored 16.6% on the test.
Why it mattersThat gives readers an early, concrete measure of how much work remains between AI-assisted chip design and a tool that reliably gets the job done.
Discuss: Is a benchmark score of 16.6% a useful early baseline for AI chip design, or too low to tell us much yet?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.