The German consortium behind the AI model Soofi S has acknowledged in version 3.0 of its tech report that test questions from the science benchmark GPQA accidentally ended up in the training data. The community caught t…
New models reset the capability and price-performance frontier. Teams re-evaluate what to build on whenever a launch shifts what's possible per dollar.
Summaries are aggregated for information only — follow the source link for the full story. Demo entries are illustrative.