返回提交历史
Modified
README.md
+2
-1
Added
docs/async_client.md
+95
-0
XFEstudio/gpt4free
Add async client docs
8321dca1
代码差异
2 个文件
+97
-1
@@ -188,7 +188,8 @@ image_url = response.data[0].url
188
188
189
189
**Full Documentation for Python API**
190
190
191
- New Client API like the OpenAI Python library: [/docs/client](/docs/client.md)
191
- New AsyncClient API from G4F: [/docs/async_client](/docs/async_client.md)
192
- Client API like the OpenAI Python library: [/docs/client](/docs/async_client.md)
192
193
- Legacy API with python modules: [/docs/legacy](/docs/legacy.md)
193
194
194
195
#### Web UI
@@ -0,0 +1,95 @@
1
# How to Use the G4F AsyncClient API
2
3
The AsyncClient API is the asynchronous counterpart to the standard G4F Client API. It offers the same functionality as the synchronous API, but with the added benefit of improved performance due to its asynchronous nature.
4
5
Designed to maintain compatibility with the existing OpenAI API, the G4F AsyncClient API ensures a seamless transition for users already familiar with the OpenAI client.
6
7
## Key Features
8
9
The G4F AsyncClient API offers several key features:
10
11
- **Custom Providers:** The G4F Client API allows you to use custom providers. This feature enhances the flexibility of the API, enabling it to cater to a wide range of use cases.
12
13
- **ChatCompletion Interface:** The G4F package provides an interface for interacting with chat models through the ChatCompletion class. This class provides methods for creating both streaming and non-streaming responses.
14
15
- **Streaming Responses:** The ChatCompletion.create method can return a response iteratively as and when they are received if the stream parameter is set to True.
16
17
- **Non-Streaming Responses:** The ChatCompletion.create method can also generate non-streaming responses.
18
19
- **Image Generation and Vision Models:** The G4F Client API also supports image generation and vision models, expanding its utility beyond text-based interactions.
20
21
22
## Using AsyncClient
23
24
### Text Completions:
25
26
You can use the ChatCompletions endpoint to generate text completions as follows:
27
28
```python
29
response = await client.chat.completions.create(
30
model="gpt-3.5-turbo",
31
messages=[{"role": "user", "content": "Say this is a test"}],
32
...
33
)
34
print(response.choices[0].message.content)
35
```
36
37
Streaming completions are also supported:
38
39
```python
40
stream = client.chat.completions.create(
41
model="gpt-4",
42
messages=[{"role": "user", "content": "Say this is a test"}],
43
stream=True,
44
...
45
)
46
async for chunk in stream:
47
if chunk.choices[0].delta.content:
48
print(chunk.choices[0].delta.content or "", end="")
49
```
50
51
### Image Generation:
52
53
You can generate images using a specified prompt:
54
55
```python
56
response = await client.images.generate(
57
model="dall-e-3",
58
prompt="a white siamese cat",
59
...
60
)
61
62
image_url = response.data[0].url
63
```
64
65
### Example usage with asyncio.gather
66
67
Start two tasks at the same time:
68
69
```python
70
import asyncio
71
72
from g4f.client import AsyncClient
73
from g4f.Provider import BingCreateImages, OpenaiChat, Gemini
74
75
async def main():
76
client = AsyncClient(
77
provider=OpenaiChat,
78
image_provider=Gemini,
79
# other parameters...
80
)
81
82
task1 = client.chat.completions.create(
83
model="gpt-3.5-turbo",
84
messages=[{"role": "user", "content": "Say this is a test"}],
85
)
86
task2 = client.images.generate(
87
model="dall-e-3",
88
prompt="a white siamese cat",
89
)
90
responses = await asyncio.gather(task1, task2)
91
92
print(responses)
93
94
asyncio.run(main())
95
```