SweetBench Pro
product
3 mentions
3 recordings
first heard Feb 2026
last heard 10 Jul
↑3 vs the 12 months before
OpenAI benchmark that was found to contain many broken tasks
Trend
Peak in the week of 7 Jul — 1 mentions across 1 series.
Moments
newest first · ▶ plays the momentNathaniel WhittemoreThe AI Daily Brief: Artificial Intelligence News and Analysis · ChatGPT Just Became a Work Agent · 1:51 · 10 Jul
…In a new report, OpenAI audited SweeBench Pro and found the benchmark to be sorely lacking.…
Sebastian RaschkaLatent Space: The AI Engineer Podcast · [LIVE] Anthropic Distillation & How Models Cheat (SWE-Bench ... · 43:48 · 26 Feb
…I think there is like a Sweetbench Pro now, which is kind of like uh…
Olivia WatkinsLatent Space: The AI Engineer Podcast · ⚡️The End of SWE-Bench Verified — Mia Glaese & Olivia ... · 1:29 · 23 Feb
…Like Super Red Pro.…
All 3 moments