File size: 2,782 Bytes
ede5b01 20b542c 6f03965 421b315 6d47000 6f03965 ba1145c 6b199b6 cac9ed0 59d5288 6f03965 843250c 7214838 665f389 fbb54b0 4bb8c98 1fd02c0 7bd5ccb 1fd02c0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
---
license: apache-2.0
base_model:
- Qwen/Qwen3-4B
pipeline_tag: text-generation
---
# Jan-Nano: An Agentic Model
[](https://github.com/menloresearch/deep-research)
<div align="center">
<img src="https://cdn-uploads.huggingface.co/production/uploads/65713d70f56f9538679e5a56/wC7Xtolp7HOFIdKTOJhVt.png" width="300" alt="Jan-Nano">
</div>
Authors: [Alan Dao](https://scholar.google.com/citations?user=eGWws2UAAAAJ&hl=en), [Bach Vu Dinh](https://scholar.google.com/citations?user=7Lr6hdoAAAAJ&hl=vi), [Thinh Le](https://scholar.google.com/citations?user=8tcN7xMAAAAJ&hl=en)
## Overview
Jan-Nano is a compact 4-billion parameter language model specifically designed and trained for deep research tasks. This model has been optimized to work seamlessly with Model Context Protocol (MCP) servers, enabling efficient integration with various research tools and data sources.
## Evaluation
Jan-Nano has been evaluated on the SimpleQA benchmark using our MCP-based benchmark methodology, demonstrating strong performance for its model size:
The evaluation was conducted using our MCP-based benchmark approach, which assesses the model's performance on SimpleQA tasks while leveraging its native MCP server integration capabilities. This methodology better reflects Jan-Nano's real-world performance as a tool-augmented research model, validating both its factual accuracy and its effectiveness in MCP-enabled environments.
## How to Run Locally
Jan-Nano is currently supported by [Jan - beta build](https://www.jan.ai/docs/desktop/beta), an open-source ChatGPT alternative that runs entirely on your computer. Jan provides a user-friendly interface for running local AI models with full privacy and control.
For non-jan app or tutorials there are guidance inside community section, please check those out! [Discussion](https://huggingface.co/Menlo/Jan-nano/discussions)
### VLLM
Here is an example command you can use to run vllm with Jan-nano
```
vllm serve Menlo/Jan-nano --host 0.0.0.0 --port 1234 --enable-auto-tool-choice --tool-call-parser hermes --chat-template ./qwen3_nonthinking.jinja
```
Chat-template is already included in tokenizer so chat-template is optional, but in case it has issue you can download the template here [Non-think chat template](https://qwen.readthedocs.io/en/latest/_downloads/c101120b5bebcc2f12ec504fc93a965e/qwen3_nonthinking.jinja)
### Recommended Sampling Parameters
- Temperature: 0.7
- Top-p: 0.8
- Top-k: 20
- Min-p: 0
### Documentation
[Setup, Usage & FAQ](https://menloresearch.github.io/deep-research/) |