đ Lucy: Edgerunning Agentic Web Search on Mobile with a 1.7B model.
Lucy is a 1.7B model designed for agentic web search and lightweight browsing. Built on Qwen3-1.7B, it combines the research capabilities of larger models and can run efficiently on mobile devices, even with only a CPU.
Authors: Alan Dao, Bach Vu Dinh, Alex Nguyen, Norapat Buppodom
đ Quick Start
Lucy is a compact yet powerful 1.7B model that specializes in agentic web search and lightweight browsing. Based on Qwen3-1.7B, it inherits the in - depth research capabilities of larger models and is optimized to run efficiently on mobile devices, even in CPU - only configurations.
⨠Features
- đ Strong Agentic Search: Empowered by MCP - enabled tools (e.g., Serper with Google Search).
- đ Basic Browsing Capabilities: Via Crawl4AI (MCP server to be released), Serper, etc.
- đą Mobile - Optimized: Lightweight enough to run on CPU or mobile devices at a decent speed.
- đ¯ Focused Reasoning: Machine - generated task vectors optimize thinking processes for search tasks.
đ Documentation
Overview
Lucy is a compact but capable 1.7B model focused on agentic web search and lightweight browsing. Built on Qwen3-1.7B, Lucy inherits deep research capabilities from larger models while being optimized to run efficiently on mobile devices, even with CPU - only configurations. We achieved this through machine - generated task vectors that optimize thinking processes, smooth reward functions across multiple categories, and pure reinforcement learning without any supervised fine - tuning.
Evaluation
Following the same MCP benchmark methodology used for [Jan - Nano](https://huggingface.co/Menlo/Jan - nano) and [Jan - Nano - 128k](https://huggingface.co/Menlo/Jan - nano - 128k), Lucy demonstrates impressive performance despite being only a 1.7B model, achieving higher accuracy than DeepSeek - v3 on [SimpleQA](https://openai.com/index/introducing - simpleqa/).
đĻ Installation
Deployment
Lucy can be deployed using various methods including vLLM, llama.cpp, or through local applications like Jan, LMStudio, and other compatible inference engines. The model supports integration with search APIs and web browsing tools through the MCP.
Deploy using VLLM
vllm serve Menlo/Lucy-128k \
--host 0.0.0.0 \
--port 1234 \
--enable-auto-tool-choice \
--tool-call-parser hermes
Or llama-server from llama.cpp
llama-server ...
Recommended Sampling Parameters
Temperature: 0.7
Top-p: 0.9
Top-k: 20
Min-p: 0.0
đ¤ Community & Support
đ License
This project is licensed under the Apache 2.0 license.
đ Citation
Paper (coming soon): Lucy: edgerunning agentic web search on mobile with machine generated task vectors.
đ Information Table
| Property |
Details |
| Model Type |
Compact 1.7B model for agentic web search and lightweight browsing |
| Base Model |
Qwen/Qwen3 - 1.7B |
| Pipeline Tag |
text - generation |
| Library Name |
transformers |
| License |
Apache 2.0 |
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
