Topic
Benchmarks
3 stories about Benchmarks that passed the news check, newest first.
Research ·
Thinking Machines' Inkling 2 Small drops 316 F-bombs across benchmark runs
An analysis by Vals AI found Inkling 2 Small swore 316 times across 2,655 benchmark runs, according to The Decoder.
The Decoder · 3 sources
Research ·
Free Chinese AI Models Are Getting Scary Good at Hacking
An experiment asked AI models to break into simulated networks.
Security Latest
Release ·
Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding
Ricursive released Ripple-65B Preview, an open-weight model that its maker describes as state of the art on agentic coding within a narrow U.S.-made open-weight category.
Hacker News: Front Page