Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,56 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-nd-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-nc-nd-4.0
|
3 |
+
datasets:
|
4 |
+
- kwaikeg/KAgentInstruct
|
5 |
+
- kwaikeg/KAgentBench
|
6 |
+
language:
|
7 |
+
- en
|
8 |
+
- zh
|
9 |
+
pipeline_tag: text-generation
|
10 |
---
|
11 |
+
|
12 |
+
|
13 |
+
KwaiAgents ([Github](https://github.com/KwaiKEG/KwaiAgents)) is a series of Agent-related works open-sourced by the [KwaiKEG](https://github.com/KwaiKEG) from [Kuaishou Technology](https://www.kuaishou.com/en). The open-sourced content includes:
|
14 |
+
|
15 |
+
1. **KAgentSys-Lite**: An experimental Agent Loop implemented based on open-source search engines, browsers, time, calendar, weather, and other tools, which is only missing the memory mechanism and some search capabilities compared to the system in the paper.
|
16 |
+
2. **KAgentLMs**: A series of large language models with Agent capabilities such as planning, reflection, and tool-use, acquired through the Meta-agent tuning proposed in the paper.
|
17 |
+
3. **KAgentInstruct**: Fine-tuned data of instructions generated by the Meta-agent in the paper.
|
18 |
+
4. **KAgentBench**: Over 3,000 human-edited, automated evaluation data for testing Agent capabilities, with evaluation dimensions including planning, tool-use, reflection, concluding, and profiling.
|
19 |
+
|
20 |
+
|
21 |
+
## User Guide
|
22 |
+
|
23 |
+
### Serving by [Lamma.cpp](https://github.com/ggerganov/llama.cpp) (CPU)
|
24 |
+
llama-cpp-python offers a web server which aims to act as a drop-in replacement for the OpenAI API. This allows you to use llama.cpp compatible models with any OpenAI compatible client (language libraries, services, etc).
|
25 |
+
|
26 |
+
To install the server package and get started:
|
27 |
+
```bash
|
28 |
+
pip install llama-cpp-python[server]
|
29 |
+
python3 -m llama_cpp.server --model kagentlms_qwen_7b_mat_gguf/ggml-model-q4_0.gguf --chat_format chatml --port 8888
|
30 |
+
```
|
31 |
+
|
32 |
+
Finally, you can use the curl command to invoke the model same as the OpenAI calling format. Here's an example:
|
33 |
+
```bash
|
34 |
+
curl http://localhost:8888/v1/chat/completions \
|
35 |
+
-H "Content-Type: application/json" \
|
36 |
+
-d '{"messages": [{"role": "user", "content": "Who is Andy Lau"}]}'
|
37 |
+
```
|
38 |
+
|
39 |
+
## Citation
|
40 |
+
```
|
41 |
+
@article{pan2023kwaiagents,
|
42 |
+
author = {Haojie Pan and
|
43 |
+
Zepeng Zhai and
|
44 |
+
Hao Yuan and
|
45 |
+
Yaojia Lv and
|
46 |
+
Ruiji Fu and
|
47 |
+
Ming Liu and
|
48 |
+
Zhongyuan Wang and
|
49 |
+
Bing Qin
|
50 |
+
},
|
51 |
+
title = {KwaiAgents: Generalized Information-seeking Agent System with Large Language Models},
|
52 |
+
journal = {CoRR},
|
53 |
+
volume = {abs/2312.04889},
|
54 |
+
year = {2023}
|
55 |
+
}
|
56 |
+
```
|