All articles tagged with #ai benchmarks
Google's Gemini 3 Flash, a more efficient version of its latest AI model, performs comparably to GPT-5.2 in benchmarks, even surpassing it in some tests, signaling a significant advancement in AI capabilities and competition with OpenAI.
The AMD Ryzen AI Max+ 'Strix Halo' SoCs successfully ran ROCm 7.0 on Ubuntu Linux, despite not being listed on the supported GPU list, with benchmarks showing functional performance on AI tasks and graphics workloads.
OpenAI has released GPT-5-Codex, an upgraded version of its AI coding agent, which features dynamic thinking capabilities allowing it to spend varying amounts of time on tasks, improving performance on coding benchmarks and code reviews. The model is now available to various ChatGPT users and will be accessible via API in the future, aiming to enhance competitiveness in the crowded AI coding market.
Google DeepMind has launched Gemini 2.5 Deep Think, its most advanced multi-agent AI reasoning model capable of exploring multiple ideas simultaneously to improve answer quality, outperforming other models on various benchmarks, and integrating tools like code execution and search. The model is available to subscribers and aims to enhance research and problem-solving capabilities, with plans for broader testing.
Google has released an upgraded preview of Gemini 2.5 Pro, enhancing coding capabilities and overall performance, with improvements in creativity and response formatting, set for general availability in a few weeks.