mem-agent

Based on Qwen3-4B-Thinking-2507, this model was trained using GSPO (Zheng et al., 2025) over an agent scaffold that is built around an Obisidian-like memory system and the tools required to interact with it. The model was trained on the following subtasks:

• Retrieval: Retrieving relevant information when needed from the memory system. In this subtask, we also trained the model on filtering the retrieved information and/or obfuscating it completely.
• Updating: Updating the memory system with new information.
• Clarification: Asking for clarification when the user query is not clear/contradicting with the information in the memory system.

The tools in the scaffold are:

# File Operations
create_file(file_path: str, content: str = "") -> bool  # Auto-creates parent directories
update_file(file_path: str, old_content: str, new_content: str) -> Union[bool, str] # Returns True or error message
read_file(file_path: str) -> str
delete_file(file_path: str) -> bool
check_if_file_exists(file_path: str) -> bool

# Directory Operations
create_dir(dir_path: str) -> bool
list_files() -> str  # Shows tree structure of current working directory
check_if_dir_exists(dir_path: str) -> bool

# Utilities
get_size(file_or_dir_path: str) -> int  # Bytes; empty = total memory size
go_to_link(link_string: str) -> bool

In the scaffold, the model uses <think>, <python> and <reply> tags to structure its response. Using <reply> only when it's done interacting with the memory. The <python> block is executed in a sandbox with the tools and the results of the code block are returned in a <result> tag to the model, forming the agentic loop.

The model is also trained to be able to handle optional filters given by the user in between tags after the user query. These filters are used to filter the retrieved information and/or obfuscate it completely.

Benchmark

We evaluated this model and a few other open & closed ones on our benchmark, md-memory-bench. We used o3 from OpenAI as the judge. All the other models except driaforall/mem-agent and Qwen/Qwen3-4B-Thinking-2507 were used through OpenRouter.s

Our model, with only 4B parameters, is only second on the benchmark, beating all the open & closed models except for qwen/qwen3-235b-a22b-thinking-2507. The model achieves an overall score of 0.75, a significant improvement over the 0.3929 of the base Qwen model.

Usage

The model, while can be used on its own, is recommended to be used as an MCP server to a bigger model, which can then be used to interact with the memory system. For this, you can check our repo, which contains instructions for both an MCP setup and a cli standalone model usage.

Memory

The model uses a markdown based memory system with links, inspired by Obsidian. The general structure of the memory is:

memory/
    ├── user.md
    └── entities/
        └── [entity_name_1].md
        └── [entity_name_2].md
        └── ...

• user.md is the main file that contains information about the user and their relationships, accompanied by links to the enity file in the format of [[entities/[entity_name].md]] per relationship. The link format should be followed strictly.
• entities/ is the directory that contains the entity files.
• Each entity file follows the same structure as user.md.
• Modifying the memory manually does not require restarting the MCP server.

Example user.md

# User Information
- user_name: John Doe
- birth_date: 1990-01-01
- birth_location: New York, USA
- living_location: Enschede, Netherlands
- zodiac_sign: Aquarius

## User Relationships
- company: [[entities/acme_corp.md]]
- mother: [[entities/jane_doe.md]]

Example entity files (jane_doe.md and acme_corp.md)

# Jane Doe
- relationship: Mother
- birth_date: 1965-01-01
- birth_location: New York, USA

# Acme Corporation
- industry: Software Development
- location: Enschede, Netherlands

The model is trained on this memory standard and any fruitful use should be on a memory system that follows this standard. We have a few memory export tools for different sources like ChatGPT, Notion, etc. in our mcp server repo.

References:

• GSPO, Zheng et al., 2025