commit 3d861ee8a10e128b9ec87a731531ca8c1337db9f Author: ModelHub XC Date: Thu Oct 8 09:51:18 2026 +0800 初始化项目,由ModelHub XC社区提供模型 Model: Menlo/Lucy-128k-gguf Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..1a94774 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,50 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q4_1.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q5_0.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q5_1.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +lucy_128k-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +lucy_demo[[:space:]](online-video-cutter.com).gif filter=lfs diff=lfs merge=lfs -text +lucy_demo.gif filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..1ef5d40 --- /dev/null +++ b/README.md @@ -0,0 +1,90 @@ +--- +license: apache-2.0 +language: +- en +base_model: +- Menlo/Lucy-128k +pipeline_tag: text-generation +library_name: transformers +--- + +# Lucy: Edgerunning Agentic Web Search on Mobile with a 1.7B model. + +[![GitHub](https://img.shields.io/badge/GitHub-Repository-blue?logo=github)](https://github.com/menloresearch/deep-research) +[![License](https://img.shields.io/badge/License-Apache%202.0-yellow)](https://opensource.org/licenses/Apache-2.0) + +
+ Lucy-128k +
+ +**Authors:** [Alan Dao](https://scholar.google.com/citations?user=eGWws2UAAAAJ&hl=en), [Bach Vu Dinh](https://scholar.google.com/citations?user=7Lr6hdoAAAAJ&hl=vi), [Alex Nguyen](https://github.com/nguyenhoangthuan99), [Norapat Buppodom](https://scholar.google.com/citations?user=utfEThsAAAAJ&hl=th&authuser=1) + + +![image/gif](lucy_demo.gif) + + +## Overview + +Lucy is a compact but capable 1.7B model focused on agentic web search and lightweight browsing. Built on [Qwen3-1.7B](https://huggingface.co/Qwen/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. + +## What Lucy Excels At + +- **🔍 Strong Agentic Search**: Powered by MCP-enabled tools (e.g., Serper with Google Search) +- **🌐 Basic Browsing Capabilities**: Through Crawl4AI (MCP server to be released), Serper,... +- **📱 Mobile-Optimized**: Lightweight enough to run on CPU or mobile devices with decent speed +- **🎯 Focused Reasoning**: Machine-generated task vectors optimize thinking processes for search tasks + +## 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/). + +![image/png](https://cdn-uploads.huggingface.co/production/uploads/65713d70f56f9538679e5a56/lG2FqLCWXq1N8lh7wlJgW.png) + +## 🖥️ How to Run Locally + +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. + +### Deployment + +Deploy using VLLM: +```bash +vllm serve Menlo/Lucy-128k \ + --host 0.0.0.0 \ + --port 1234 \ + --enable-auto-tool-choice \ + --tool-call-parser hermes \ + --rope-scaling '{"rope_type":"yarn","factor":3.2,"original_max_position_embeddings":40960}' --max-model-len 131072 +``` + +Or `llama-server` from `llama.cpp`: +```bash +llama-server ... --rope-scaling yarn --rope-scale 3.2 --yarn-orig-ctx 40960 +``` + +### Recommended Sampling Parameters + +```yaml +Temperature: 0.7 +Top-p: 0.9 +Top-k: 20 +Min-p: 0.0 +``` + +## 🤝 Community & Support + +- **Discussions**: [HuggingFace Community](https://huggingface.co/Menlo/Lucy-128k/discussions) + +## 📄 Citation +```bibtex +@misc{dao2025lucyedgerunningagenticweb, + title={Lucy: edgerunning agentic web search on mobile with machine generated task vectors}, + author={Alan Dao and Dinh Bach Vu and Alex Nguyen and Norapat Buppodom}, + year={2025}, + eprint={2508.00360}, + archivePrefix={arXiv}, + primaryClass={cs.CL}, + url={https://arxiv.org/abs/2508.00360}, +} +``` +**Paper **: *Lucy: edgerunning agentic web search on mobile with machine generated task vectors.* \ No newline at end of file diff --git a/lucy_128k-Q3_K_L.gguf b/lucy_128k-Q3_K_L.gguf new file mode 100644 index 0000000..4ca35a2 --- /dev/null +++ b/lucy_128k-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aee86ada61137081dcdc4e8ec131c54be6c1b17eb79f65e282a8e044d19be165 +size 1003502848 diff --git a/lucy_128k-Q3_K_M.gguf b/lucy_128k-Q3_K_M.gguf new file mode 100644 index 0000000..99c31f7 --- /dev/null +++ b/lucy_128k-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2db05d455b3fe48880bd9f3f67dfbadbba7174313d0ee5d325401904adeae09b +size 939539712 diff --git a/lucy_128k-Q3_K_S.gguf b/lucy_128k-Q3_K_S.gguf new file mode 100644 index 0000000..6791ae4 --- /dev/null +++ b/lucy_128k-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc3622b400abfc2f39f296a19e9e571dc4952918401a6a8e84c00e308c9956d6 +size 867253504 diff --git a/lucy_128k-Q4_0.gguf b/lucy_128k-Q4_0.gguf new file mode 100644 index 0000000..4dea679 --- /dev/null +++ b/lucy_128k-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1acbda87424507cf4ec20a3489469a5baf71bf0c152b063d377ea40f9acf50d8 +size 1054424320 diff --git a/lucy_128k-Q4_1.gguf b/lucy_128k-Q4_1.gguf new file mode 100644 index 0000000..3e31e3a --- /dev/null +++ b/lucy_128k-Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cceddb038c28934d5dd58bfcd454cf9498761663eb71a6dbb2c64241275155b6 +size 1142504704 diff --git a/lucy_128k-Q4_K_M.gguf b/lucy_128k-Q4_K_M.gguf new file mode 100644 index 0000000..cc0cad6 --- /dev/null +++ b/lucy_128k-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf959f8c3855a6104b53f835fc0d93e6756c80d524cc92dadef5bb886d3e18ed +size 1107410176 diff --git a/lucy_128k-Q4_K_S.gguf b/lucy_128k-Q4_K_S.gguf new file mode 100644 index 0000000..12eacc0 --- /dev/null +++ b/lucy_128k-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d6ec369c6fcc009aaaa40fdf59e974579a4cae0a0d58d4467042653120ea6d3 +size 1060191488 diff --git a/lucy_128k-Q5_0.gguf b/lucy_128k-Q5_0.gguf new file mode 100644 index 0000000..549da63 --- /dev/null +++ b/lucy_128k-Q5_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ace602ea2eb5efcfe0933826b1daf9ebd231736267bfe30621581a8cc51c1c0f +size 1230585088 diff --git a/lucy_128k-Q5_1.gguf b/lucy_128k-Q5_1.gguf new file mode 100644 index 0000000..d15c034 --- /dev/null +++ b/lucy_128k-Q5_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5a8bccdbd58883e8710a79c5bb3ba230f9a447e4e80eef04e43e74fae624b3b +size 1318665472 diff --git a/lucy_128k-Q5_K_M.gguf b/lucy_128k-Q5_K_M.gguf new file mode 100644 index 0000000..98b611e --- /dev/null +++ b/lucy_128k-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2242b371432508e50ae65b2b6cf55dd766166d90d7742f4d8048afb743bcc0e2 +size 1257880832 diff --git a/lucy_128k-Q5_K_S.gguf b/lucy_128k-Q5_K_S.gguf new file mode 100644 index 0000000..4a022ae --- /dev/null +++ b/lucy_128k-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05191e0287add196067daaeb13732db978a7e292dabaf5db50e571a3184af51f +size 1230585088 diff --git a/lucy_128k-Q6_K.gguf b/lucy_128k-Q6_K.gguf new file mode 100644 index 0000000..e0cc415 --- /dev/null +++ b/lucy_128k-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac72e2f4225becf7c3d382fa357a5d6fe8386e04a51dd4f2c6e84d1faee1424a +size 1417755904 diff --git a/lucy_128k-Q8_0.gguf b/lucy_128k-Q8_0.gguf new file mode 100644 index 0000000..ff2e306 --- /dev/null +++ b/lucy_128k-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:55d28ecf761da3c5b73b69a053d57bdd1fdb1f4903ac7070b7800aa9d6dda601 +size 1834427648 diff --git a/lucy_demo.gif b/lucy_demo.gif new file mode 100644 index 0000000..74b82d1 --- /dev/null +++ b/lucy_demo.gif @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:433d2004e90db167890f4d70a5b98116e05290125abbf665e7d1aede3641a24c +size 250222647