At its Ignite conference in San Francisco on Tuesday, Microsoft announced a number of new AI models, including two new The post Microsoft’s new coding model promises near-Opus performance at 5% of...
Production traffic shows failures that benchmarks or pre-deployment evaluations can’t fully reveal. It tells you which edge cases matter and The post CoreWeave connects custom training loops to...
绿盟SecLLM-3.6-27B在CyberGym纯模型榜单以91.31%的任务成功率位居全球第一,较此前54.51%提升36.8个百分点。该榜单仅使用通用框架、不接入外部工具,衡量模型自身的安全能力。模型采用安全能力拆解训练,经持续预训练、长链路轨迹与多教师蒸馏,强化漏洞理解、定位、PoC构造与失败恢复能力。选择27B规模是为实现低成本本地化部署,目前已在告警降噪、漏洞初筛等场景落地。
Explore a candidate nonperturbative lattice definition of the Standard Model and how GPT-6 Astra helped turn the idea into a detailed proof.
Google’s new Gemini Decisions API, introduced Wednesday, is designed to handle the routine decisions that AI applications make thousands of The post Google built an AI model that can’t write a...
Anthropic just released a coding model that beats Opus 5.5 on two major benchmarks and can complete certain difficult tasks The post Anthropic built a model that beats Opus. It recommends Opus...
Docusign rolled out AI coding tools across its engineering teams, expecting them to help developers write code faster and do The post Moving from AI adoption to an AI operating model: How Docusign...
Cognition built its newest coding model on a foundation that anyone can download, yet the company is keeping its own The post Cognition turned an open model into a coding beast. But it won’t...
Starting on Friday, everyone using Meta’s Muse agent and Meta AI gets new models the company trained using Muse Wonder, The post Meta’s best AI model now trains the one most of its users get...
Anthropic has published how the Claude Code team reviews code now that Claude writes roughly 90% of it, including a claim that nobody keeps a complete mental model of the codebase. The post...
Datadog has released Scout, a model trained to investigate production problems. It runs inside Bits Investigation, the company's AI SRE product, where it can pick up an alert, look through...
Banks increasingly deploy AI in distressed credit management. But few have adjusted their operating models to deliver distinctive customer experiences and optimize financial outcomes and recovery costs.
Authors: Divya Gadde, Ko-Jen Hsiao, Dhaval Patel and Moumita BhattacharyaThe motivation for building the Netflix Foundation Model for personalization was laid out in this previous blog post. The...
Google has introduced support for local models in the Antigravity SDK, enabling developers to build and execute agentic workflows directly on their own machines, while retaining the flexibility to...
本文提出HAT训练法,用于解决直播数字人Agent在低延迟下适应业务规则频繁变化的问题。该方法将动态执行环境Harness状态纳入训练分布,分三阶段:HSA-SFT提升工具调用能力,通用指令蒸馏恢复通用能力,HSA-RL在仿真环境学习真实交互。基于Qwen3.6-35B-A3B的模型在直播问答达94.8分,超越千亿参数大模型,Harness变化下保持94.6分,端到端延迟P50仅3.407秒,已落地淘宝直播数字人业务。
特斯拉将Powershare家庭备用电源功能扩展至Model 3和Model Y,此前仅限Cybertruck。该功能需搭配Powerwall 3使用,可为家庭供电两天以上,但仅保证2025年10月1日后订购的车辆支持,且暂限美国及波多黎各。
Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.
TypeSafe AI, founded by former OpenAI researcher Diogo Almeida, has introduced Jev, a decision-making model that generates typed outputs rather than text. It evaluates inputs in parallel,...
Google has published benchmarks for GKE Pod snapshots, reporting up to 89% lower startup latency and a 70B model loading in 37 seconds. The feature checkpoints CPU and GPU memory through gVisor...
OpenAI has released a disclosure framework for model misalignment during its lifecycle. Employees can flag potential issues, prompting technical staff to label incidents. The initial case studies...
完成下面两步后,将自动完成登录并继续当前操作。