Jim Fan's team open-sources the agentic robotics framework ASPIRE V1.0
Jim Fan's team has open-sourced the agentic robotics framework ASPIRE V1.0, which features real-robot skill discovery and self-improvement capabilities. The team says that after switching backend models on the day Astra launched, all benchmarks reached about 99% overnight. In subsequent comparisons, Opus 5.5 slightly outperformed Astra on robotics simulation benchmarks, while Sonnet 5.5 landed on the Pareto frontier of cost versus robotics capability at $12 per task. ASPIRE and the /robot skill library were first launched in June.
Why read this
AI score 64
ASPIRE V1.0 is open-sourced and delivers real-robot skill discovery and self-improvement capabilities, along with comparison data of different backend models on robotics benchmarks.
Topics
Body
We are open-sourcing V1.0 of our agentic robotics framework, ASPIRE, with full real robot skill discovery and self improvement capabilities! On the day Astra arrived, our lab switched the backend model and saturated all the benchmarks to ~99% over night.
And the frontier is already on the move: - Opus 5.5 slightly edges out Astra on our robot simulation bench. - Sonnet 5.5 hits the Pareto frontier of cost vs. robot capability, $12/task.
We first introduced ASPIRE and /robot skill library in June, way before the big models could reliably do robotics. A rising frontier tide lifts all boats, but Astra was a 0->1 tsunami on many tasks.
Now you can also feel the Physical Bitter Lesson at home like we did on that fateful day ;)
links in thread