OpenAI Unveils New AI Tools for Enhanced Collaboration and Customization
OpenAI announced new AI tools during DevDay 2026, expanding ChatGPT's capabilities for developers and users with features like plugins and agents, targeting 1.2B weekly users.
Model Release
576 articles
New AI models as they ship — large language models, multimodal systems, and open-weight releases. Capabilities, context windows, pricing, and availability, summarized so you can compare what is actually new.
LATEST NEWS
-
-
Anthropic Releases Claude Sonnet 5.5 with 30 Percent Lower Cost
Anthropic's Claude Sonnet 5.5 nearly matches Opus 5.5 on benchmarks while costing up to 30 percent less per task, according to thedecoder.
-
AWS Launches Claude Sonnet 5.5 for Enhanced Coding and Knowledge Work
AWS announced the availability of Claude Sonnet 5.5 on Amazon Bedrock today, offering faster task completion and lower costs for focused coding and knowledge work.
-
OpenAI Focuses 80-90% of Research on GPT-7 and Beyond
OpenAI's Head of Applied Research, Boris Power, said 80 to 90 percent of the company's research now targets GPT 7 and beyond, focusing on long-term value.
-
Nvidia Releases Free 100M-Parameter Model for Real-Time Speaker Identification
Nvidia announced a free 100M-parameter model that identifies up to eight speakers in real time, outperforming its predecessor by 41% on a benchmark test.
-
Meta Launches Early Access for New Muse Features
Meta is offering early access to its Muse AI app's latest features, with users able to request participation starting September 25, 2026.
-
AMD Introduces Model Weight Profiles for Llama-3.1
AMD released model weight profiles on September 25, 2026, showing how parameters are distributed across embeddings, attention, and dense layers in Llama-3.1 models.
-
Google Launches Call for Me with Gemini to Handle Phone Calls for Users
Google is testing 'Call for Me,' a feature that lets Gemini call businesses on users' behalf, available to Pixel 11 owners in the US with a paid subscription.
-
Astra and Opus Break Enigma Codes Using AI
Two AI models, Astra and Opus, cracked two long-unsolved Enigma messages, one since 2005, using advanced decryption techniques.
-
Meta's Muse AI Agent Outpaces ChatGPT in Early Testing
Meta's Muse AI agent reportedly outperformed ChatGPT's early mobile launch, with early numbers suggesting it could be a major contender in the AI space.
-
AWS Deploys Qwen3-TTS Model for Real-Time Voice Cloning
AWS announced the deployment of the Qwen3-TTS-12Hz-1.7B-Base model on Amazon SageMaker AI, enabling real-time personalized speech synthesis with a short reference recording.
-
Google Deepmind's New Chief Prioritizes Gemini 4 Over AGI
Google Deepmind's new chief, Koray Kavukcuoglu, aims to release Gemini 4 before year-end, shifting focus from AGI to practical product development.