From 0d1ae405cc24f3daafb783d758c019ef6731d3aa Mon Sep 17 00:00:00 2001
From: Heiner Lohaus <heiner@lohaus.eu>
Date: Thu, 26 Oct 2023 21:32:49 +0200
Subject: Add Llama2 Providers / Models

---
 g4f/Provider/DeepInfra.py | 63 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 g4f/Provider/DeepInfra.py

(limited to 'g4f/Provider/DeepInfra.py')

diff --git a/g4f/Provider/DeepInfra.py b/g4f/Provider/DeepInfra.py
new file mode 100644
index 00000000..70045cae
--- /dev/null
+++ b/g4f/Provider/DeepInfra.py
@@ -0,0 +1,63 @@
+from __future__ import annotations
+
+import json
+from aiohttp import ClientSession
+
+from ..typing import AsyncResult, Messages
+from .base_provider import AsyncGeneratorProvider
+
+
+class DeepInfra(AsyncGeneratorProvider):
+    url = "https://deepinfra.com"
+    working = True
+
+    @classmethod
+    async def create_async_generator(
+        cls,
+        model: str,
+        messages: Messages,
+        proxy: str = None,
+        **kwargs
+    ) -> AsyncResult:
+        if not model:
+            model = "meta-llama/Llama-2-70b-chat-hf"
+        headers = {
+            "User-Agent": "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/118.0",
+            "Accept": "text/event-stream",
+            "Accept-Language": "de,en-US;q=0.7,en;q=0.3",
+            "Accept-Encoding": "gzip, deflate, br",
+            "Referer": f"{cls.url}/",
+            "Content-Type": "application/json",
+            "X-Deepinfra-Source": "web-page",
+            "Origin": cls.url,
+            "Connection": "keep-alive",
+            "Sec-Fetch-Dest": "empty",
+            "Sec-Fetch-Mode": "cors",
+            "Sec-Fetch-Site": "same-site",
+            "Pragma": "no-cache",
+            "Cache-Control": "no-cache",
+        }
+        async with ClientSession(headers=headers) as session:
+            data = {
+                "model": model,
+                "messages": messages,
+                "stream": True,
+            }
+            async with session.post(
+                "https://api.deepinfra.com/v1/openai/chat/completions",
+                json=data,
+                proxy=proxy
+            ) as response:
+                response.raise_for_status()
+                first = True
+                async for line in response.content:
+                    if line.startswith(b"data: [DONE]"):
+                        break
+                    elif line.startswith(b"data: "):
+                        chunk = json.loads(line[6:])["choices"][0]["delta"].get("content")
+                        if chunk:
+                            if first:
+                                chunk = chunk.lstrip()
+                                if chunk:
+                                    first = False
+                            yield chunk
\ No newline at end of file
-- 
cgit v1.2.3