Interesting new benchmark: https://x.com/serenaa_ge/status/2059308218564890875

Seems like a good benchmark. What about adding this to OpenHands Index?

I have an audit of the benchmark, please skim or point your agent at it: https://www.june.kim/auditing-deepswe

FYI - Customers asked me about DeepSWE benchmark and I also noticed Meta/Muse Spark and X/Grok included it in their release

june.kim/auditing-deepswe-v1-1 They fixed a lot of the issues in v1.1 I expect that in 4 months or so they'll release a V2