The feature allows the agent to run multiple tasks at once while users control login choices and approve downloads.
Using modified training protocols on Qwen3.6-35B-A3B and gpt-oss-20b, the research demonstrates that models can perform serially dependent reasoning steps without visible chain-of-thought output.
Opus 5.5 is now the default model on Pro, Max, Ultra, and Team plans, offering performance comparable to Opus 5 at High effort while using 20 to 25% fewer output tokens.
The feature is rolling out on Team and Enterprise plans for Google Meet, with notes and transcripts free during the beta period.
The model succeeds Gemini 3.8 Flash, handling longer engineering tasks and producing better-researched documents in fewer agent steps.
The guide uses a coffee-ordering API example to show how to define tasks, run trials in Vercel Sandboxes, and analyze traces to fix agent failures.
The study shows that best-of-N alignment closely approaches the theoretical maximum reward gain, while PPO and GRPO remain substantially suboptimal.
The plugin connects to Autodesk Fusion, Infor, Microsoft 365, and Oracle NetSuite to draft quotes and reports for human review.
A guide based on interviews with 8 startups shows how Claude Tag automates triage, PRs, and data queries in Slack, with Anthropic internally answering 75% of data questions in 1-2 minutes.
A new script and container allow users to fine-tune TRELLIS.2 with LoRA on custom 3D assets, significantly improving geometric accuracy for specific domains like footwear.
Built with Claude Fable 5.1, Opus 5.5, Sonnet 5.5, and Haiku 5.5, the tool is available now for women's and men's styles in the U.S.
The funding supports a scientific research network under the Coalition for Children’s Rights and Protection in the Age of AI to develop tools for measuring AI risks.
The new benchmark uses 330+ error types to weight parsing errors by severity, correcting previous metrics that penalized formatting over factual errors.
BankSallad Financial Services is recruiting the second cohort of financial solution experts to build an AI OS-based sales model.
Apiiro identified 104,185 public instances of .claude/settings.local.json, with 1,122 containing exposed secrets due to missing project-level gitignore rules.
Nvidia's board is reportedly preparing to oust CEO Jensen Huang following allegations that he had an affair with a woman U.S. intelligence agencies believe was a Chinese asset.
Loudness and age tags significantly influence listener preference, while usage metrics and professional status show no correlation with quality.
The author argues that AI success has shifted from 75% skills to 55% skills and 45% connections due to the commoditization of intelligence, high resource requirements, and reduced observability.
The open-source library supports exact k-NN, k-means, NN-Descent, and CAGRA algorithms, outperforming CPU-based hnswlib in both build time and search speed.
MACaQ quantizes a 122B-parameter model in 64 minutes using 32.5 GB of device memory, supporting text, image, video, and audio models across llama.cpp, vLLM, Diffusers, and PyTorch.
The carrier cites risks of data exposure, unauthorized mile usage, and regulatory issues with automated check-ins.
The update introduces Team Bots with role-based access, shared connectors, and the ability to join Google Meet, Zoom, or Microsoft Teams calls for note-taking.
The tool runs on Linux, macOS, and Windows, supports AWS SDKs and CLI, and is built in Rust to minimize resource usage.
The study argues that defining slop by community rejection allows for measurable governance of AI-generated content.
AI Briefing