143 stories
MiniMax has announced M2.1, an advanced open-source AI model specializing in multilingual programming languages such as Rust, Java, and Go, achieving a leading 72.5% score on the SWE-bench Multilingual benchmark and surpassing competitors like Gemini 3 Pro and Claude Sonnet 4.5.
December 23, 2025
December 23, 2025
December 22, 2025
December 20, 2025
Tech • December 12, 2025
OpenAI's GPT-5.2 has claimed the top spot on the GDPval-AA benchmark, surpassing Claude Opus 4.5, but its high operational cost raises questions about its economic viability.
Tech • December 12, 2025
OpenAI has unveiled GPT-5.2, featuring three specialized variants with significant improvements in benchmarks, long-context reasoning, and reduced hallucinations, eliciting mixed reactions from users on its enhanced performance and tone.
Tech • December 9, 2025
Industry reports suggest OpenAI is preparing to release GPT-5.2 this week, following an internal “code red” directive to improve ChatGPT. The update is expected to boost reasoning, accuracy, and overall stability amid rising pressure from Google’s Gemini 3.
Tech • December 8, 2025
xAI has officially released Grok 4.1, bringing notable upgrades in accuracy, conversational flow, emotional understanding, and performance benchmarks. The new model is now available across grok.com, mobile apps, and X.
Tech • December 8, 2025
Google officially launched Gemini 3.0 on November 18, 2025, presenting its most advanced AI model with deeper reasoning, enhanced multimodal capabilities, and direct integration into core Google services.