Zhipu AI Unveils GLM-5.3, Claims Strongest Open-Weights Coding Model
Zhipu AI released GLM-5.3, claiming it is the strongest open-weights coding model, with improvements focused on agent-based tasks.
Browse all published articles.
Zhipu AI released GLM-5.3, claiming it is the strongest open-weights coding model, with improvements focused on agent-based tasks.
Anthropic's Claude Code generated 388 pull requests for software maintenance, with 180 merged at a 46 percent rate, according to internal testing.
Joi AI paid 10 consultants $2,000 each to masturbate daily for a month, tracking effects on stress, focus, and cravings.
HuggingFace's AX-Ray identifies causal-leakage defects in two public models, highlighting a structural correctness issue that impacts deployment safety.
Suno Studio 2.0 introduces a chat feature allowing users to interact with the AI music platform as if it were a bandmate, with no credit cost for plugin creation.
Meta CEO Mark Zuckerberg released a 6,500-word AI manifesto, but critics argue it lacks concrete substance and fails to address key industry concerns.
IBM announced a partnership with OpenAI to integrate the latter’s AI models into its consulting services, aiming to enhance enterprise AI adoption.
Deepseek released its updated V4-Pro model, which scores higher on agent benchmarks but still trails top models like Claude Opus 5. The company also open-sourced its agent software and raised API prices.
Google launched Gemini 3.7 Flash, a coding-focused model with 50% lower pricing than its predecessor, three weeks after Gemini 3.6 Flash.
Google has launched Gemini 3.7 Flash, a new version of its large language model, with improved coding performance and a lower introductory price.
Instagram unveiled a redesigned wordmark after 10 years, with critics noting its AI-like appearance despite no direct AI use in the design process.
OpenAI has replaced its chief revenue officer after nine months, appointing Wiz’s Dali Rajic to lead sales as the company continues reshuffling its executive team.