update readme

This commit is contained in:
Wendong
2025-03-09 13:59:44 +08:00
parent 9f77090c4a
commit 041d2644dd
2 changed files with 34 additions and 6 deletions

View File

@@ -154,6 +154,21 @@ Run the following demo case:
python owl/run.py
```
## Running with Different Models
OWL supports various LLM backends. You can use the following scripts to run with different models:
```bash
# Run with Qwen model
python owl/run_qwen.py
# Run with Deepseek model
python owl/run_deepseek.py
# Run with other OpenAI-compatible models
python owl/run_openai_compatiable_model.py
```
For a simpler version that only requires an LLM API key, you can try our minimal example:
```bash
@@ -169,7 +184,7 @@ question = "Task description here."
society = construct_society(question)
answer, chat_history, token_count = run_society(society)
logger.success(f"Answer: {answer}")
print(f"Answer: {answer}")
```
For uploading files, simply provide the file path along with your question:
@@ -180,8 +195,7 @@ question = "What is in the given DOCX file? Here is the file path: tmp/example.d
society = construct_society(question)
answer, chat_history, token_count = run_society(society)
logger.success(f"Answer: {answer}")
print(f"Answer: {answer}")
```
OWL will then automatically invoke document-related tools to process the file and extract the answer.

View File

@@ -154,6 +154,21 @@ python owl/run.py
python owl/run_mini.py
```
## 使用不同的模型
OWL 支持多种 LLM 后端。您可以使用以下脚本来运行不同的模型:
```bash
# 使用 Qwen 模型运行
python owl/run_qwen.py
# 使用 Deepseek 模型运行
python owl/run_deepseek.py
# 使用其他 OpenAI 兼容模型运行
python owl/run_openai_compatiable_model.py
```
你可以通过修改 `run.py` 脚本来运行自己的任务:
```python
@@ -163,7 +178,7 @@ question = "Task description here."
society = construct_society(question)
answer, chat_history, token_count = run_society(society)
logger.success(f"Answer: {answer}")
print(f"Answer: {answer}")
```
上传文件时,只需提供文件路径和问题:
@@ -175,12 +190,11 @@ question = "给定的 DOCX 文件中有什么内容文件路径如下tmp/e
society = construct_society(question)
answer, chat_history, token_count = run_society(society)
logger.success(f"答案:{answer}")
print(f"答案:{answer}")
```
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
你可以尝试以下示例任务: