summaryrefslogtreecommitdiffstats
path: root/g4f/Provider/DeepInfra.py
diff options
context:
space:
mode:
authorH Lohaus <hlohaus@users.noreply.github.com>2024-04-07 11:27:26 +0200
committerGitHub <noreply@github.com>2024-04-07 11:27:26 +0200
commitd327afc60620913f5d2b0a9985b03a7934468ad4 (patch)
tree395de9142af3e6b9c0e5e3968ee7f8234b8b25e2 /g4f/Provider/DeepInfra.py
parentUpdate Gemini.py (diff)
parentUpdate provider.py (diff)
downloadgpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar.gz
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar.bz2
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar.lz
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar.xz
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.tar.zst
gpt4free-d327afc60620913f5d2b0a9985b03a7934468ad4.zip
Diffstat (limited to 'g4f/Provider/DeepInfra.py')
-rw-r--r--g4f/Provider/DeepInfra.py70
1 files changed, 20 insertions, 50 deletions
diff --git a/g4f/Provider/DeepInfra.py b/g4f/Provider/DeepInfra.py
index 6cf52694..53c8d6b9 100644
--- a/g4f/Provider/DeepInfra.py
+++ b/g4f/Provider/DeepInfra.py
@@ -1,42 +1,41 @@
from __future__ import annotations
-import json
import requests
from ..typing import AsyncResult, Messages
-from .base_provider import AsyncGeneratorProvider, ProviderModelMixin
-from ..requests import StreamSession, raise_for_status
+from .needs_auth.Openai import Openai
-class DeepInfra(AsyncGeneratorProvider, ProviderModelMixin):
+class DeepInfra(Openai):
url = "https://deepinfra.com"
working = True
+ needs_auth = False
supports_stream = True
supports_message_history = True
default_model = 'meta-llama/Llama-2-70b-chat-hf'
-
+
@classmethod
def get_models(cls):
if not cls.models:
url = 'https://api.deepinfra.com/models/featured'
models = requests.get(url).json()
- cls.models = [model['model_name'] for model in models]
+ cls.models = [model['model_name'] for model in models if model["type"] == "text-generation"]
return cls.models
@classmethod
- async def create_async_generator(
+ def create_async_generator(
cls,
model: str,
messages: Messages,
stream: bool,
- proxy: str = None,
- timeout: int = 120,
- auth: str = None,
+ api_base: str = "https://api.deepinfra.com/v1/openai",
+ temperature: float = 0.7,
+ max_tokens: int = 1028,
**kwargs
) -> AsyncResult:
headers = {
'Accept-Encoding': 'gzip, deflate, br',
'Accept-Language': 'en-US',
'Connection': 'keep-alive',
- 'Content-Type': 'application/json',
+ 'Content-Type': None,
'Origin': 'https://deepinfra.com',
'Referer': 'https://deepinfra.com/',
'Sec-Fetch-Dest': 'empty',
@@ -44,46 +43,17 @@ class DeepInfra(AsyncGeneratorProvider, ProviderModelMixin):
'Sec-Fetch-Site': 'same-site',
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36',
'X-Deepinfra-Source': 'web-embed',
- 'accept': 'text/event-stream',
+ 'Accept': None,
'sec-ch-ua': '"Google Chrome";v="119", "Chromium";v="119", "Not?A_Brand";v="24"',
'sec-ch-ua-mobile': '?0',
'sec-ch-ua-platform': '"macOS"',
}
- if auth:
- headers['Authorization'] = f"bearer {auth}"
-
- async with StreamSession(headers=headers,
- timeout=timeout,
- proxies={"https": proxy},
- impersonate="chrome110"
- ) as session:
- json_data = {
- 'model' : cls.get_model(model),
- 'messages': messages,
- 'temperature': kwargs.get("temperature", 0.7),
- 'max_tokens': kwargs.get("max_tokens", 512),
- 'stop': kwargs.get("stop", []),
- 'stream' : True
- }
- async with session.post('https://api.deepinfra.com/v1/openai/chat/completions',
- json=json_data) as response:
- await raise_for_status(response)
- first = True
- async for line in response.iter_lines():
- if not line.startswith(b"data: "):
- continue
- try:
- json_line = json.loads(line[6:])
- choices = json_line.get("choices", [{}])
- finish_reason = choices[0].get("finish_reason")
- if finish_reason:
- break
- token = choices[0].get("delta", {}).get("content")
- if token:
- if first:
- token = token.lstrip()
- if token:
- first = False
- yield token
- except Exception:
- raise RuntimeError(f"Response: {line}")
+ return super().create_async_generator(
+ model, messages,
+ stream=stream,
+ api_base=api_base,
+ temperature=temperature,
+ max_tokens=max_tokens,
+ headers=headers,
+ **kwargs
+ ) \ No newline at end of file