is this some kind of reverse marketing jiu-jitsu?
12:05 PM · Jun 30, 2026 · 122 Views
Samuel Albanie, Google DeepMind's Gemini frontier evals lead, questions the cost-performance scaling of Claude models on agentic benchmarks
Opus 4.8 reached an 83.5% pass rate on OSWorld-Verified.
--0--
Sentiment
Sentiment building, check back later.
Cluster Engagement
-
Views
-
Comments
-
Reposts
-
Bookmarks
Expand data
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
VIEWS730LIKES7
🍓🍓🍓@iruletheworldmo
@SamuelAlbanie i can’t figure out if i’m missing something
2hViews 730Likes 7
yau@neurasleaze
@SamuelAlbanie Maybe they want to push more users to opus for better inference batching? Or the numbers are warped by subsidized API costs. Maybe they subsidize opus more than sonnet
1hViews 50