Releases

OpenAI launched web search in ChatGPT with timely answers and source links.

Oracle introduced a generative-AI clinical workflow and documentation agent.

GitHub introduces Spark, an AI-native tool for building application experiences from natural language.

Midjourney launched Retexture for changing a scene's materials, lighting, and surfaces.

Midjourney launched editing for uploaded external images.

Cohere Labs released Aya Expanse 32B as the larger Aya Expanse multilingual model.

Cohere Labs released Aya Expanse 8B with improved multilingual generation across 23 languages.

Gradio 5 introduced redesigned application theming, stronger security defaults, and improved AI app development workflows.

Hugging Face launched HUGS, optimized zero-configuration deployment containers for open generative models.

Runway released Act-One for expressive performance capture and character animation.

Anthropic launched computer use in public beta, allowing developers to direct Claude to operate graphical computer interfaces.

Anthropic released an upgraded Claude 3.5 Sonnet with major coding and tool-use improvements.

IBM released its Granite 3.0 8B model with enterprise and agentic capabilities under Apache 2.0.

IBM released its compact Granite 3.0 2B enterprise language model under Apache 2.0.

Dow Jones and the New York Post sued Perplexity, alleging copyright and trademark violations involving publisher content.

Suno partnered with Audible Magic on content identification and rights-management technology.

IBM released Docling for converting complex business documents into structured AI-ready formats.

DeepSeek released Janus 1.3B for multimodal understanding and image generation.

DeepSeek introduced the Janus family for unified multimodal understanding and generation.

Meta released research on a multimodal language model that freely mixes text and speech.

Perplexity launched Spaces and Internal Knowledge Search for organizing research and searching company files alongside the web.

Suno launched Scenes for creating soundtracks from photos and videos.

Mistral AI launched the Ministral 3B and 8B models for edge, on-device and local use.

Perplexity introduced dedicated finance capabilities for stocks, company earnings and market research.

Anthropic updated its Responsible Scaling Policy with revised capability thresholds, safety-case processes, and governance measures.

CoreWeave secured a $650 million revolving credit facility from a syndicate of major banks to support global expansion.

Amazon introduced generative-AI shopping guides that summarize category information and organize product recommendations.

Anthropic launched asynchronous batch processing for large volumes of Claude API requests.

Meta introduced foundation models for video generation, personalized video, editing and synchronized audio.

OpenAI introduced Canvas, a collaborative interface for writing and coding projects in ChatGPT.

OpenAI closed a $6.6 billion funding round at a $157 billion post-money valuation.

The LeRobot community introduced the SO-100, a low-cost open robot arm design for data collection and imitation learning.

OpenAI launched the Realtime API public beta for low-latency multimodal experiences.

Meta published improved SAM 2.1 checkpoints and developer tooling.

Oracle released a managed retrieval-augmented generation and enterprise-agent service.

Meta released a 90B multimodal model for image understanding.

Meta released an 11B multimodal model for image understanding.

Meta released a compact 1B text model for on-device use.

Meta released a compact 3B text model for on-device use.

OpenAI began expanding ChatGPT's Advanced Voice Mode with more natural real-time audio conversations.

Palantir expanded its Maven Smart System work with the U.S. government under a five-year agreement with a potential value up to $100 million.

Anthropic introduced Contextual Retrieval, combining contextual embeddings and BM25 to improve retrieval-augmented generation.

Modal added secure tunnels and execution improvements for sandbox workloads.

Modal added improved object-storage bucket mounting for data-intensive jobs.

Modal expanded secret management for applications and integrations.

Modal expanded queue-based communication between serverless functions.

Modal expanded its persistent key-value storage capabilities.

Modal improved persistent volume performance and reliability.

Modal expanded persistent storage and data-access capabilities for cloud workloads.

Modal improved support for distributed and GPU-intensive training workloads.