Anthropic has a 2-hour engineering take-home test. It says its new Claude 4.5 model outscored every human who took it.

Anthropic is solidifying its dominance in AI coding, with a new release that performed better on its engineering test than ...
Ars Technica

A new open-weights AI coding model is closing in on proprietary options

On Tuesday, French AI startup Mistral AI released Devstral 2, a 123 billion parameter open-weights coding model designed to work as part of an autonomous software engineering agent. The model achieves ...
ZDNet

I tested Opus 4.5 to see if it's really 'the best in the world' at coding - and things got weird fast

Opus 4.5 failed half my coding tests, despite bold claims File handling glitches made basic plugin testing nearly impossible Two tests passed, but reliability issues still dominate the story I've got ...
Geeky Gadgets

DeepSWE AI Coding Model Benchmark Finally Solves AI Training Data Contamination

DeepSWE, created by DataCurve offers a benchmark for assessing AI coding models by focusing on real-world programming challenges rather than synthetic test cases. According to Matthew Berman, one of ...