Grok 4.5 Climbs to #2 on FrontierSWE

elonmusk · x · 2026-07-15

A repost claims Grok 4.5 has risen to #2 on the FrontierSWE benchmark.

The post emphasizes that this makes Grok 4.5 one of the strongest software engineering models, showing its continued improvement on real-world coding tasks and agentic development workflows.

Related event: Grok 4.5 Rises to No. 2 on FrontierSWE(5 posts)→

Original post →

More from coding & agent

coding & agent channel →