rsi-benchmark
RSI Bench is an ongoing effort to evaluate whether AI agents can develop the capabilities required to advance AI R&D.
scaleapi/rsi-benchmark
View on GitHub ↗RSI Bench is an ongoing effort to evaluate whether AI agents can develop the capabilities required to advance AI R&D.
rsi-benchmark ranks 398th of 440 in Research & Deep Research, actively maintained, last commit today.
Star trajectory
29 days tracked2026-09-2598 stars today2026-10-24
Momentum
- in 30d, below the Research & Deep Research median of +0.7%.
Maintenance
Last commit today, fresher than the median of 40.5d.
How it compares in Research & Deep Research
rsi-benchmark
98
stars
- 30d
commit today
Research & Deep Research median
274.5
stars
+0.7% 30d
commit ~40.5d ago
Leader · autoresearch
98.5k
stars
+1.9% 30d
commit 212 days ago
Ranked 398th of 440 by stars.
- Language
- Python
- License
- MIT License
- Contributors
- 21
- Forks
- 53
- Watchers
- 0
- Open issues
- 4
- Releases
- 0
- Created
- 2026-07-30
- Stars this week
- +12
You've seen where rsi-benchmark stands today.
Get the moves before everyone else: the 5 Research & Deep Research repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.