Open Source news

DeepSeek says Harness has 1.54 million daily users

DeepSeek’s open-source agent app went live on August 13. The company says it now uses about 92 trillion tokens a day. No outside group has checked the numbers. DeepSeek also plans an official marketplace for the app’s plugins.

DeepSeek says its agent app Harness had 1.54 million daily active users as of November 4. The count covers only people who use it through DeepSeek’s own API. At the end of September, the figure was about 271,000, Bloomberg and Yicai report.

Torvalds says AI-found fixes made Linux 7.4 extra big

The two weeks for new code brought almost 20,000 commits, against about 15,000 usually. Torvalds calls it a “much healthier model” that AI-found fixes now arrive early, not during testing.

Linus Torvalds released the first test version of Linux 7.4, called rc1, on Sunday, November 1. He wrote that this was a “BIG” merge window, with almost 20,000 commits, and he linked it to “AI triggered work”. He named a second reason too: many expect 7.4 to be a long-term support version.

DeepSeek open-sources DeepInfer, the engine behind its API

The free code runs DeepSeek V4.1 Flash on Nvidia’s Blackwell chips and on Huawei’s Ascend 950. Some parts that DeepSeek uses for its own service are still missing.

DeepSeek has published DeepInfer, the software it says runs its own large-scale AI service, under the Apache 2.0 licence. For now it supports one model, DeepSeek V4.1 Flash, on two kinds of chips: Nvidia’s SM100 GPUs and Huawei’s Ascend 950.

Microsoft’s MXC sandbox for AI agents reaches version 1.0

Microsoft Execution Containers keep AI agents inside limits on Windows 11, macOS and Linux. The open-source kit left preview on October 7, and GitHub Copilot and OpenAI’s Codex already use it, Microsoft says.

Microsoft made its Execution Containers (MXC) generally available on October 7, four months after a first preview in June. Version 1.0 of the free, MIT-licensed kit has stable programming interfaces for Rust, .NET and Node.js. It lets developers decide which files and actions an AI agent may reach.

Anthropic offers free AI bug scans to open-source projects

Reports go straight to maintainers who opt in, with no human check first. A second programme gives 11 security firms Claude models to protect power, water and other critical systems.

Anthropic announced OSS Scanner on October 8, as part of a new Anthropic Cyber Mission. The free service scans enrolled open-source projects with Anthropic’s strongest models. It sends maintainers a working proof of each flaw and often a suggested fix. Only core maintainers can enrol, and Anthropic checks each one by hand.

Google pauses a bug bounty, flooded by automated reports

Its open-source reward program stopped taking reports of flaws in its code on October 1. Google says most automated submissions were not valid, and promises an update in early 2027.

Google has paused the main part of its Open Source Software Vulnerability Reward Program. Since October 1, it no longer accepts reports of security flaws in the code of its open-source projects. Google blames a sharp rise in automated reports, most of which were not valid.

Pop!_OS maker bans AI-generated code from its COSMIC desktop

System76 now rejects LLM-written code, comments and descriptions in pull requests and issues across its Pop!_OS projects. Reports say AI-made submissions were swamping reviewers.

System76, which makes the Linux system Pop!_OS and its COSMIC desktop, changed its contribution rules on September 30. Every pull request must now confirm it contains no content generated by an LLM, “including code, comments, and descriptions”. Pull requests without the completed checklist will be closed.

Meta lets hobbyists build their own Muse gadgets

Meta released open-source firmware and a Linux kit that connect cheap boards to its Muse agent. It is also giving away 5,000 Muse Home Link dongles to US subscribers.

Meta launched Muse Gadgets on October 2, an open-source project for building small devices that talk to Muse, its personal AI agent. The firmware and software kits run on ESP32 boards and the Raspberry Pi 5, under the Apache 2.0 licence. Meta also offers a free USB-C Muse Home Link that lets Muse control devices on a home network.

DeepSeek open-sources tools for Huawei’s AI chips

Its engineers added Huawei’s Ascend 950 to TileLang, an open-source language for AI chip code. DeepSeek says Huawei gave full support.

DeepSeek said on September 30 that it is open-sourcing programming tools for Huawei’s Ascend AI chips, including compute and communication libraries. Its engineers built support for the Ascend 950 into TileLang, an open-source language for writing the code that runs on AI chips. Reuters and the South China Morning Post describe it as a step away from Nvidia’s software.

Google’s Intrinsic made its core robot software open source

The AI models and cloud tools stay paid.

Intrinsic published Intrinsic Core on GitHub on September 22, at the ROSCon conference in Toronto. The free toolkit covers real-time control, finding parts with cameras, motion and grasp planning, and simulation for robot arms. Its Flowstate tool, advanced AI models, and cloud services remain paid.

A plugin flaw hit four AI coding agents

Two vendors fixed it, and two pushed back.

Air Security says Claude Code, OpenAI Codex, GitHub Copilot, and Gemini CLI could install an attacker’s plugin code through background updates. Anthropic and OpenAI fixed it. Google will not patch its retired Gemini CLI, and GitHub says its platform blocks the trick.

Nvidia added software for fault-tolerant quantum computers

The hardware is still the hard part.

CUDA-Q Logical links quantum programs, error-correction methods, and hardware estimates. Fermilab reports faster design work, while Sandia’s QUOPS benchmark offers a common test for future systems.

Nvidia PAIR lets AI work across several computers at home

The free, open-source tool sends separate Ollama and LM Studio tasks to trusted computers on one local network. It does not combine their memory into one larger GPU.

Small open AI models are easier to use in real products

Many teams now choose a smaller model that fits their budget, device, and privacy rules. The best test score is not always the best choice.