How product and program managers can pick the right frontier AI model: cost per task, benchmarks, access rules, and when a ...
An enhanced PointNet++ with attention mechanisms and model scaling reaches up to 96.26 percent accuracy in separating ground ...
OrcaRouter's OrcaCyber Zero 1.5 is a gated 1M-context cybersecurity model scoring 100% on Cybench, 95.8% on CVE-Bench subset.
This report describes examples of unintended model actions we’ve observed during evaluations and internal use of Claude. It is part of our effort to publish more frequent standalone reports on model ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results