资讯与观察· 模型发布
AI 行业动态、工程实践、产品方法和社区故事。
Kyutai CADRE: Video Autoencoders That Cut World Model Lag Below 230 ms
Kyutai released CADRE, a family of 7 open causal diffusion video autoencoders for interactive world models. World models built on CADRE run at 20 fps on 1 H100 and react in under 230 ms.
Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context
Moonshot AI released Kimi K3.1 Pro, a 2.8T-parameter, 104B-active MoE model with 1M-token context. It scores 65.7% on Terminal-Bench 4.0, costs $2/$10 per million tokens, and open weights arrive by Nov 4, 2026.
13 分钟 · MarkTechPost
Rapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)
Learn how OpenAI scaled Habitat with reliable multi-tenancy, global caching, and Azure Cosmos DB to deliver 12× faster reads for ChatGPT.
30 分钟 · OpenAILATEST STORIES
最新内容
LitPivot: Developing research ideas through dynamic contextualization within the literature landscape精选导读
LitPivot: Developing research ideas through dynamic contextualization within the literature landscape
Securing our frontier model training environment精选导读
OpenAI explains how we strengthened containment for frontier model training and evaluation, including offline environments and high-compute red-teaming.
TinyFish Breaks Down Agentic Search Architecture for Production AI Agents精选导读
TinyFish published two engineering research articles on agentic search. One defines a 7-phase production architecture with evidence contracts and bounded retries. The other explains why its deep research system dropped …
Google Cloud Launches Agent Compute Engine: A MicroVM per AI Agent精选导读
Google Cloud announced Agent Compute Engine, a compute service that gives every AI agent its own hardware-isolated microVM with no credentials inside. It already runs Gemini agents and enters preview in the coming month…
You.com Launches Inference Router: 12 Open-Weight Models With 99.9% Uptime SLA精选导读
You.com launched Inference Router, an OpenAI-compatible API for 12 open-weight models with a 99.9% uptime SLA, zero data retention, 100% US hosting, server-side web search and no platform fee.
OpenAI and Poland’s NASK partner to advance AI safety and security精选导读
OpenAI and Poland’s NASK are partnering on AI safety, security, and technical research to strengthen safeguards and support a more resilient digital ecosystem.
Does Your Model Run on AWS Trainium? Look It Up精选导读
Does Your Model Run on AWS Trainium? Look It Up
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox精选导读
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
Build more responsive apps with the Decisions API精选导读
Fast, focused, low-cost decisions for any application.