FM News
Benchmarking Opus 5 on SlopCodeBench
On a story from Hacker News3w ago
Read the original at github.com
Benchmarking Opus 5 on SlopCodeBench
fmode.me/n/benchmarking-opus-5-on-slopcodebench
More from FM News
Quoting OpenClaw1 pts · Simon WillisonCoding agents can be evaluated. We just have to evaluate the work.1 pts · The New StackAI coding got faster. Why didn’t engineering?1 pts · The New StackThe playbook for building high talent density teams | Adam Ward, Head of Talent at Cursor1 pts · Lenny's Newsletter