Simplex AI releases Linger code for local personal agents
Read more
Pandaily
pandaily.com

Simplex AI releases Linger code for local personal agents

Simplex AI has made the source code for Linger public—a personal agent designed to run on a user's own computer using local language models. The project was featured in Chinese publications on November 20th. The project founder, Tony Zhang, previously led the DeepSeek-VL project.

Linger is a product of Simplex AI, not a release from DeepSeek. The system's central agent, named Another Me, is designed to structure information from emails, calendars, chats, and meetings into a unified user knowledge base. It can also identify tasks requiring follow-up and delegate them to other agents. The project description also mentions browser and desktop operation capabilities with the option for human intervention when necessary. It should be noted that these functions are described by the developers and have not undergone independent product testing.

The local inference component, called PolyStrata, distributes the loads of 'mixture-of-experts' type models across the Graphics Processing Unit (GPU) memory, System Random Access Memory (RAM), and SSD storage, instead of requiring all experts to reside on the video card. More frequently used experts remain closer to the GPU, while less demanded parts can be served through other memory levels.

Simplex AI reports that the Qwen3.8-Flash-Next model with 125 billion parameters generates 45.5 tokens per second on an RTX 3060 with 12GB of VRAM and 64GB of RAM, and 179 tokens per second on an RTX 5090 with 32GB. These figures are manufacturer-reported decoding speed measurements and do not represent full agent task time characteristics or guarantees of equivalent quality to cloud models. Additional hardware planning tables include budget memory configuration simulations, which limits comparisons between different machines.

The public release of the code does not imply unlimited licensing. The main license modifies Apache 2.0 by adding additional conditions: running Linger as a hosted or managed service for third parties requires written permission from the manufacturer, and removing or changing the frontend branding is only possible with a commercial license. The terms allow for internal commercial use and use as a backend for the organization's own applications.

At the time of checking, the main website indicated that downloads for macOS, Windows, and Linux as ready-made packages are expected in the future, despite the fact that the source code and setup documentation were publicly available. Code availability should not be confused with the release of a finished installer. Furthermore, local inference does not guarantee fully autonomous operation, as connected applications, voice commands, remote models, and background tasks using external subscriptions may transmit data to external providers. The security documentation describes server assertions and checks, not a sandbox isolating agents from owner files. Pandaily did not install the software nor reproduce the claimed performance.

Similar stories

Federal Revenue Service makes software codes available on GitHub to aid tax reform
Read more
tecnoblog.net

Federal Revenue Service makes software codes available on GitHub to aid tax reform

The Federal Revenue Service has made its software projects public for the first time in open repositories on GitHub. This action includes the Tax Calculator, a tool designed to apply the rules of the consumption tax reform and can be integrated into business systems.

According to a statement issued by the agency this Monday, November 16th, the projects were released under the MIT license. This license allows the code to be used, modified, and distributed, including in commercial contexts. The disclosure of this measure was formalized in the Official Gazette of the Union.

The Tax Calculator incorporates the calculation guidelines for the Contribution on Goods and Services (CBS), the Tax on Goods and Services (IBS), and other taxes provided for in the reform. The purpose is to enable programmers to use the same logic employed by the Revenue Service in their own billing, management, and fiscal document issuance programs.

The agency made a total of three projects available on GitHub. The Federal Revenue Service expects this initiative to help reduce discrepancies in calculations, development costs, and difficulties encountered in adapting to the new regulations.

With these projects on GitHub, developers gain the ability to check different versions, monitor any updates, and examine in detail how taxes are calculated. The MIT license also offers flexibility to adapt the codes to various applications.

The Revenue Service assures that this move aims to enhance both transparency and legal certainty during the implementation of the tax reform. It is important to note that the use of open-source software does not imply the absence of rules; even permissive licenses, such as the MIT, impose specific conditions for the reuse and redistribution of the projects.

Although the calculator was already available for download on the CBS Platform since 2025, the novelty now lies in its publication on GitHub, an environment that facilitates version tracking and code analysis by the developer community. The projects were developed in collaboration with Serpro and do not aim to replace the systems currently used by companies, but rather to provide official components that can be inserted into existing solutions during the transition to the new tax regime.

AgentOS from openJiuwen allows agents to independently create, install, and roll back plugins
Read more
pandaily.com

AgentOS from openJiuwen allows agents to independently create, install, and roll back plugins

openJiuwen, an open agent community supported by Huawei, has introduced a mechanism it calls RSI-native AgentOS. This system is an extension that allows AI agents to add functionality to their own application, test it in real operation, and roll it back if errors occur. The term RSI stands for recursive self-improvement.

The mechanism provides extension points in five different parts of the stack. Client extensions add buttons, panels, and settings to the web interface. Gateway extensions are responsible for request routing and authentication. AgentServer extensions process server commands, messages, and sessions. Swarm extensions configure how multiple agents distribute work and track progress. Agents can adjust prompts and register tools, MCP services, or sub-agents.

New features are released as extension packages, each containing a manifest.json file. This manifest declares the identifier, version, dependencies, and code entry point, and also specifies where any interface component should appear. Upon installation, the system verifies the manifest and saves the package by its identifier and version. Then, the Gateway and AgentServer load the corresponding packages and register handlers in their event buses, which trigger upon events such as button clicks, incoming requests, or the agent assembling a system prompt. These handlers are capable of calling operations provided by each component, such as reading session data or updating the interface.

The packages are hot-pluggable, meaning adding a new one does not require changing the application's source code. In the WorkSwarm agent application developed by openJiuwen, this mechanism is used for so-called system plugins. The user describes the desired function, and the agent writes, packages, and installs it autonomously. In demonstrations, openJiuwen WorkSwarm added a 'pet' mode to the desktop, which collapses the workstation into a small chat companion, as well as 'Token Weather'—a status panel displaying session token usage as weather. If a plugin causes a failure, users have the option to restore an earlier checkpoint of their plugin's configuration. Plugins can also be installed manually from a folder, the extensions page, or using the /extension install command.

A similar mechanism is applicable to enterprise needs. openJiuwen demonstrates an entry plugin for the Gateway, adapters for third-party developer sandboxes, and a plugin allowing sessions to run on Claude Code, Codex, or within the openJiuwen environment. In distributed deployment, the Gateway and AgentServer can function independently, with each side loading its own extensions.

The WorkSwarm code is available on GitHub under the name JiuwenSwarm and is distributed under the Apache-2.0 license. openJiuwen characterizes this as part of its Hyper-Agent architecture and names this transition a shift in focus from developer-friendly design to RSI-oriented design.

China introduces AI agent identification registry in Uzhgen; 20 organizations sign self-discipline pact
Read more
pandaily.com

China introduces AI agent identification registry in Uzhgen; 20 organizations sign self-discipline pact

At the forum in Uzhgen dedicated to trust, interconnectivity, and agent compatibility, a platform for registering and resolving AI agent identification data was presented on November 4th. This was reported by IT Home and the Chinese website China Daily. As part of the same event, 20 companies, including Huawei, Alibaba, Tencent, Xiaomi, and Baidu, adopted a convention on self-discipline in the development of AI agents.

The platform is currently under development. According to available data, it is being created jointly by state structures to ensure identifier coding, cross-domain resolution, and reliable compatibility. Additionally, operational environments will be placed on top of the platform to host access and publish agent capabilities. A pilot program has already been launched, connecting several service providers and operating platforms.

Ant Group was one of the first participants. The company integrated its agent collaboration infrastructure into the operating platform. Early services, such as the Alipay and Sesame Credit government service agents, established and completed trusted connections with agents on other platforms. Chen Liang, Vice President of Ant Group and CTO of its Alipay and Security business group, emphasized that the age of agents requires a trust infrastructure that embeds identification, authorization, and accountability into every interaction. He also mentioned Ant's unified identification account APASS and the ASL protocol, which the company is developing with over 20 partners.

The convention was developed by the Chinese Academy of Information and Communications Technology (CAICT) together with companies such as Huawei, Alibaba, Tencent, Xiaomi, Baidu, Volcano Engine, Ant Group, and Douyin. It includes eight commitments concerning agent trust systems, interconnectivity infrastructure, compatibility, security protection, management mechanisms, user rights, industry implementation, and open technology exchange. Signatories pledged to develop compliance verification and rule embedding technologies, as well as jointly combat the use of agents for fraud and algorithmic exploitation.

The infrastructure section requires establishing common standards for agent registration and resolution, capability discovery, intent translation, routing, and protocol compatibility. The forum was hosted by the Cyberspace Administration of China and GSMA. Wang Jintao, Deputy Director of the Cyberspace Administration of China, called for establishing basic rules for agent identification, trusted interconnectivity, authorization management, and appropriate operation, expanding online governance from content and model security to verifiable subjects, controlled behavior, and traceable accountability. Parties that signed the convention stated their readiness to accept supervision from industry bodies and the public.

Popular