AI — Page 66

AI coding agents have turned GitHub reliability into an infrastructure problem Azure cannot absorb alone on Microsoft's timetable.

Teknium says the delegate tool can now fan out work without freezing the chat, a practical change for long-running agent workflows.

Karan Goel is using benchmark wins to pitch Cartesia as both the speaking and listening layer for real-time AI agents.

CrankGPT runs speech recognition, a small language model, and text-to-speech locally on a Raspberry Pi 5 with no battery or cloud.

A 73-year-old developer said he mentioned feeling unwell during a coding task, and Claude Code kept urging immediate care before doctors treated a sudden AFib episode.

Z Lab, Modal and LMSYS released a DFlash drafter for Qwen's 397B model and benchmarked it above native MTP on 8x B200 GPUs.

The company pulled Fable 5 and Mythos 5 after a June 12th export-control order, then sent technical staff to Washington to repair the relationship.

Zohar Alon's new identity-security startup is betting enterprises will need to govern agents like workers, not service accounts.

The updated model card says a base merge of Nex-N2-Pro and Qwen was uploaded by mistake, shifting the dispute from pure attribution to release discipline.

Alexandr Wang's Muse Spark gives Meta a proprietary model; the harder job is proving it can become more than ad infrastructure.

A June preprint claims Pearl's GPU network is doing random matrix math, not verified AI work, challenging Omri Weinstein's core bet.

Yichuan Wang and collaborators show a screenshot-first retrieval system beating text pipelines, with lower agent token use and a real chunking gap.

Przemek Chojecki's 14-problem smoke run puts Moonshot's new open-weight model behind Claude Fable-5-max and ahead of GPT-5.5 xhigh.

Matei Zaharia and Kasey Uhlenhuth's alpha project sits above Claude Code, Codex, the Pi agent and custom agents to compose multi-agent workflows, share live sessions and enforce contextual policies.

IplanRIO put Rio 3.5 Open 397B on Hugging Face with MIT licensing, a 1M-token context claim, and self-reported gains over Qwen's base model.

Meta's 6,500-person Applied AI unit looks less like automation magic than the unglamorous human operating system required to make models useful.

The new flagship adds High and Max reasoning modes and a 1M-context configuration for coding agents such as Claude Code and OpenClaw.

The open-source coding agent says Claude Fable 5 planned better, while GPT-5.5 matched it on execution at lower cost.

Anthropic pulled Fable 5 and Mythos 5 after a June 12th U.S. directive, turning its safety-first model rollout into a policy test case.

The directive reaches foreign nationals inside the US, including Anthropic employees, turning frontier-model access into an export-compliance problem.

Zyphra is pairing open weights, Apache 2.0 licensing, hosted inference, and its own TTS eval in a direct challenge to closed voice platforms.

Vercel's changelog says AI SDK 7 adds `HarnessAgent`, an experimental canary API for running Claude Code, Codex and Pi through sandboxed sessions and SDK-compatible streams.

The paper tested OpenEvidence and UpToDate Expert AI against GPT-5.2, Gemini 3.1 Pro and Claude Opus 4.6 across three medical evaluations.

Kimi.ai says Kimi-K2.7-Code beats K2.6 on coding benchmarks and is meant for Kimi Code and the Kimi API, while a new beta program will give applicants early access to upcoming models and features.