Commit
·
b1d0895
1
Parent(s):
38255bb
update readme
Browse files- README.md +17 -3
- README_zh.md +17 -3
README.md
CHANGED
@@ -154,6 +154,21 @@ Run the following demo case:
|
|
154 |
python owl/run.py
|
155 |
```
|
156 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
157 |
For a simpler version that only requires an LLM API key, you can try our minimal example:
|
158 |
|
159 |
```bash
|
@@ -169,7 +184,7 @@ question = "Task description here."
|
|
169 |
society = construct_society(question)
|
170 |
answer, chat_history, token_count = run_society(society)
|
171 |
|
172 |
-
|
173 |
```
|
174 |
|
175 |
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
|
|
180 |
|
181 |
society = construct_society(question)
|
182 |
answer, chat_history, token_count = run_society(society)
|
183 |
-
|
184 |
-
logger.success(f"Answer: {answer}")
|
185 |
```
|
186 |
|
187 |
OWL will then automatically invoke document-related tools to process the file and extract the answer.
|
|
|
154 |
python owl/run.py
|
155 |
```
|
156 |
|
157 |
+
## Running with Different Models
|
158 |
+
|
159 |
+
OWL supports various LLM backends. You can use the following scripts to run with different models:
|
160 |
+
|
161 |
+
```bash
|
162 |
+
# Run with Qwen model
|
163 |
+
python owl/run_qwen.py
|
164 |
+
|
165 |
+
# Run with Deepseek model
|
166 |
+
python owl/run_deepseek.py
|
167 |
+
|
168 |
+
# Run with other OpenAI-compatible models
|
169 |
+
python owl/run_openai_compatiable_model.py
|
170 |
+
```
|
171 |
+
|
172 |
For a simpler version that only requires an LLM API key, you can try our minimal example:
|
173 |
|
174 |
```bash
|
|
|
184 |
society = construct_society(question)
|
185 |
answer, chat_history, token_count = run_society(society)
|
186 |
|
187 |
+
print(f"Answer: {answer}")
|
188 |
```
|
189 |
|
190 |
For uploading files, simply provide the file path along with your question:
|
|
|
195 |
|
196 |
society = construct_society(question)
|
197 |
answer, chat_history, token_count = run_society(society)
|
198 |
+
print(f"Answer: {answer}")
|
|
|
199 |
```
|
200 |
|
201 |
OWL will then automatically invoke document-related tools to process the file and extract the answer.
|
README_zh.md
CHANGED
@@ -154,6 +154,21 @@ python owl/run.py
|
|
154 |
python owl/run_mini.py
|
155 |
```
|
156 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
157 |
你可以通过修改 `run.py` 脚本来运行自己的任务:
|
158 |
|
159 |
```python
|
@@ -163,7 +178,7 @@ question = "Task description here."
|
|
163 |
society = construct_society(question)
|
164 |
answer, chat_history, token_count = run_society(society)
|
165 |
|
166 |
-
|
167 |
```
|
168 |
|
169 |
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
|
|
175 |
society = construct_society(question)
|
176 |
answer, chat_history, token_count = run_society(society)
|
177 |
|
178 |
-
|
179 |
```
|
180 |
|
181 |
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
|
182 |
|
183 |
-
|
184 |
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
|
185 |
|
186 |
你可以尝试以下示例任务:
|
|
|
154 |
python owl/run_mini.py
|
155 |
```
|
156 |
|
157 |
+
## 使用不同的模型
|
158 |
+
|
159 |
+
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:
|
160 |
+
|
161 |
+
```bash
|
162 |
+
# 使用 Qwen 模型运行
|
163 |
+
python owl/run_qwen.py
|
164 |
+
|
165 |
+
# 使用 Deepseek 模型运行
|
166 |
+
python owl/run_deepseek.py
|
167 |
+
|
168 |
+
# 使用其他 OpenAI 兼容模型运行
|
169 |
+
python owl/run_openai_compatiable_model.py
|
170 |
+
```
|
171 |
+
|
172 |
你可以通过修改 `run.py` 脚本来运行自己的任务:
|
173 |
|
174 |
```python
|
|
|
178 |
society = construct_society(question)
|
179 |
answer, chat_history, token_count = run_society(society)
|
180 |
|
181 |
+
print(f"Answer: {answer}")
|
182 |
```
|
183 |
|
184 |
上传文件时,只需提供文件路径和问题:
|
|
|
190 |
society = construct_society(question)
|
191 |
answer, chat_history, token_count = run_society(society)
|
192 |
|
193 |
+
print(f"答案:{answer}")
|
194 |
```
|
195 |
|
196 |
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
|
197 |
|
|
|
198 |
OWL 将自动调用与文档相关的工具来处理文件并提取答案。
|
199 |
|
200 |
你可以尝试以下示例任务:
|