DeepSeek's Zhang Bo Unveils New Startup: An Open-Source Personal Agent

AI AgentDeepSeekMoE Inference EngineLocal LLMPersonal AssistantOpen Source
1 hour agoSource: blockweeks.com
DeepSeek's Zhang Bo Unveils New Startup: An Open-Source Personal Agent

Gathered, we are a fire; scattered, we are stars across the sky.

Just now, the author of

DeepSeek
V1 and the project lead of

DeepSeek
-VL, Tony Zhang (Zhang Bo), has officially announced his new startup project: Linger.

The direction is exactly the hottest thing right now: the personal Personal Agent.

The approach is also quite interesting. While peers are all making apps, they chose the computer first.

This means Linger can operate the webpages, inbox, and calendar already on your computer, without the hassle of a whole bunch of login credentials.

When it gets stuck somewhere, you can also directly take over.

Another point: as a middle layer, Linger supports connecting to various third-party models.

And it has been specifically optimized for local LLMs.

Specifically, a single RTX 3060 graphics card with 12GB of VRAM can run a 125B-parameter Qwen3.8-Flash-Next, at a generation speed of 45.5 tokens/second.

DeepSeek

Oh right, there's also something quite important—

This system is open source...

As for why?

In the announcement post, Tony Zhang cited the history of the rise of the PC:

In the 1960s, computing was centralized, until the PC put power into everyone's hands. Today, AI is trained in data centers and then rented to you in the form of tokens.

We believe intelligence should be distributed. Your context, your data, and the intelligence that drives them should be entirely owned by you.

So today, they open-sourced Linger, a dedicated Agent that lives on your personal computer.

DeepSeek

It also continues the fine tradition of the big fat whale 😭.

An AI that "takes initiative and can self-learn"

The core of Linger is an Agent system called "Another Me".

First of all, it is a memory hub.

Linger will locally read your inbox, calendar, chat records, and meetings, and every night automatically organize this information into a personal knowledge base (Wiki) for all Agents to query.

In the LoCoMo benchmark, which measures long-term conversational memory capability, Linger's accuracy reached 95%, surpassing multiple memory-enhanced AI systems such as Mem0 and MemOS.

DeepSeek

With these, Linger can work proactively better.

It will quietly observe your information stream and decide when to remind you based on the importance and urgency of events. If it receives a client email on Monday requesting a quote by Friday, it will not disturb you immediately, but wait until Wednesday morning to pop up a reminder.

But if it is a meeting time change, it will notify you immediately.

DeepSeek

This Agent also learns from your feedback, which Linger calls "implicit reinforcement".

For example, when it reminds you that an invoice is about to expire, and you reply, "I received it last week, you should have told me then." The system will record a clue.

If the next time another invoice has a similar situation and you express the same preference again, the system will automatically generate a rule: "Supplier invoice → remind immediately".

DeepSeek

Conversely, if you show disinterest in "delivery reminders" twice in a row (for example, by directly ignoring them or replying "don't tell me"), it will also learn to stay silent in the future.

You can also give direct instructions to teach it all at once. For example, tell it: "Notify me immediately about customer replies," or "After 9 PM, there's no need to tell me about meeting changes right away."

This continuous interaction and learning makes the Agent increasingly understand your work habits, becoming a truly personalized assistant.

In addition, Linger can also:

Operate the desktop and browser.

As mentioned at the beginning, Linger can directly use the Chrome and various apps already logged in on your computer, with no need to reconfigure.

Build an Agent team.

This is the Super Agent principle that

DeepSeek
explained earlier.

When you assign a complex task, Another Me will call or create Agents with different functions to collaborate on the work.

And you only need to stay in this one Session with Linger.

Voice interaction.

This mainly serves office scenarios.

Linger can transcribe meetings in real time, and after the meeting you can directly ask the Agent about the meeting content and have it follow up on to-dos.

In short, it is a Personal Agent optimized specifically for office scenarios.

However, some netizens do not quite understand this product positioning:

Why not develop based on the DeepSeek harness 🤣 What's the difference from ds-bot?

In response, Tony Zhang replied:

Linger's focus is on optimizing the model and harness together: a 12GB graphics card can run a 125B model locally, with no per-token fees, and data never leaves the local machine.

From a product positioning perspective: the core of DS-Bot is collaboration between Bots and Bots, and between Bots and groups; the core of Linger is between a person and their own Agent, and between the Agent and the local environment.

DeepSeek

12GB graphics card, runs a 125B model

To enable consumer-grade hardware to run hundred-billion-scale large models, Linger's core technology is an MoE inference engine called PolyStrata.

Its principle can be understood as an intelligent tiered storage mechanism.

For a huge MoE (Mixture of Experts) model, PolyStrata places the most commonly used and most active "expert modules" in the fastest GPU memory, puts the less commonly used ones in memory, and places the least frequently used ones on the SSD solid-state drive.

When the model needs to call a certain expert, the system dynamically moves data between the GPU, memory, and SSD. In this way, there is no need to stuff the entire model of tens or hundreds of GB into limited GPU memory.

The effect is immediate.

According to official data, a 125B-parameter Qwen3.8-Flash-Next model can achieve a generation speed of 45.5 tokens/second on a computer equipped with a 12GB RTX 3060 and 64GB of memory.

On the more powerful RTX 5090 (32GB), the speed soars to 179 tokens/second.

To make it more convenient for users to use local large models, Linger has also created a "Model Autopilot" feature.

It automatically detects your hardware configuration (GPU, GPU memory, memory), and then recommends and installs the most powerful model for you from combinations of 49 language models, two inference engines, multiple quantization schemes, and different context lengths.

DeepSeek

According to SimplexAI's scoring system, local devices with different configurations can achieve capabilities comparable to mainstream cloud large models:

8GB graphics card + 32GB memory: score 35.4, roughly equivalent to Gemini 2.5 Pro.

Mac 24GB unified memory: score 51.9, roughly equivalent to Claude Opus 4.5.

12GB graphics card + 32GB memory: score 57.3, roughly equivalent to Claude Opus 4.6.

This means that your gaming laptop or MacBook will have a truly powerful AI brain that is on standby 24/7, has no token fees, and keeps data entirely local.

DeepSeek-VL lead starts another business

Behind Linger is a team with a deep technical background, and the resume of its founder Tony Zhang is particularly impressive.

(After all, it's

DeepSeek
-related...)

DeepSeek

Tony Zhang received his bachelor's degree from Zhejiang University, earned his PhD from the Hong Kong University of Science and Technology in 2019, and subsequently joined Microsoft Research Asia (MSRA), where he has long been engaged in computer vision and generative model research. He is the author of several well-known projects such as VQ-Diffusion and Rodin 3D digital humans.

In 2023, he joined

DeepSeek
as an AI scientist.

Here, he not only participated in the research and development of the early language model

DeepSeek
LLM V1 series
, but also served as the Project Lead of the vision-language large model

DeepSeek
-VL
. This model has been highly praised for its ability to understand images, documents, and charts in real-world scenarios.

After leaving

DeepSeek
, Tony Zhang joined Zhejiang University as a faculty member, serving as a Hundred Talents Program researcher and doctoral supervisor, and founded Simplex AI.

Linger is the latest work of Simplex AI.

DeepSeek

And as the technical core of Linger, that inference engine built specifically for local deployment?

According to Tony Zhang, it was built by an 18-year-old genius in the team.

18 years old? Just graduated from high school?

This article is from the WeChat public account "Quantum Bit", author: Jay