from __future__ import annotations
from aiohttp import ClientSession
import json
from ..typing import AsyncResult, Messages, ImageType
from ..image import to_data_uri
from .base_provider import AsyncGeneratorProvider, ProviderModelMixin
from .helper import format_prompt
class DeepInfraChat(AsyncGeneratorProvider, ProviderModelMixin):
url = "https://deepinfra.com/chat"
api_endpoint = "https://api.deepinfra.com/v1/openai/chat/completions"
working = True
supports_stream = True
supports_system_message = True
supports_message_history = True
default_model = 'meta-llama/Meta-Llama-3.1-70B-Instruct'
models = [
'meta-llama/Meta-Llama-3.1-405B-Instruct',
'meta-llama/Meta-Llama-3.1-70B-Instruct',
'meta-llama/Meta-Llama-3.1-8B-Instruct',
'mistralai/Mixtral-8x22B-Instruct-v0.1',
'mistralai/Mixtral-8x7B-Instruct-v0.1',
'microsoft/WizardLM-2-8x22B',
'microsoft/WizardLM-2-7B',
'Qwen/Qwen2-72B-Instruct',
'microsoft/Phi-3-medium-4k-instruct',
'google/gemma-2-27b-it',
'openbmb/MiniCPM-Llama3-V-2_5', # Image upload is available
'mistralai/Mistral-7B-Instruct-v0.3',
'lizpreciatior/lzlv_70b_fp16_hf',
'openchat/openchat-3.6-8b',
'Phind/Phind-CodeLlama-34B-v2',
'cognitivecomputations/dolphin-2.9.1-llama-3-70b',
]
model_aliases = {
"llama-3.1-405b": "meta-llama/Meta-Llama-3.1-405B-Instruct",
"llama-3.1-70b": "meta-llama/Meta-Llama-3.1-70B-Instruct",
"llama-3.1-8B": "meta-llama/Meta-Llama-3.1-8B-Instruct",
"mixtral-8x22b": "mistralai/Mixtral-8x22B-Instruct-v0.1",
"mixtral-8x7b": "mistralai/Mixtral-8x7B-Instruct-v0.1",
"wizardlm-2-8x22b": "microsoft/WizardLM-2-8x22B",
"wizardlm-2-7b": "microsoft/WizardLM-2-7B",
"qwen-2-72b": "Qwen/Qwen2-72B-Instruct",
"phi-3-medium-4k": "microsoft/Phi-3-medium-4k-instruct",
"gemma-2b-27b": "google/gemma-2-27b-it",
"minicpm-llama-3-v2.5": "openbmb/MiniCPM-Llama3-V-2_5", # Image upload is available
"mistral-7b": "mistralai/Mistral-7B-Instruct-v0.3",
"lzlv-70b": "lizpreciatior/lzlv_70b_fp16_hf",
"openchat-3.6-8b": "openchat/openchat-3.6-8b",
"phind-codellama-34b-v2": "Phind/Phind-CodeLlama-34B-v2",
"dolphin-2.9.1-llama-3-70b": "cognitivecomputations/dolphin-2.9.1-llama-3-70b",
}
@classmethod
def get_model(cls, model: str) -> str:
if model in cls.models:
return model
elif model in cls.model_aliases:
return cls.model_aliases[model]
else:
return cls.default_model
@classmethod
async def create_async_generator(
cls,
model: str,
messages: Messages,
proxy: str = None,
image: ImageType = None,
image_name: str = None,
**kwargs
) -> AsyncResult:
model = cls.get_model(model)
headers = {
'Accept-Language': 'en-US,en;q=0.9',
'Cache-Control': 'no-cache',
'Connection': 'keep-alive',
'Content-Type': 'application/json',
'Origin': 'https://deepinfra.com',
'Pragma': 'no-cache',
'Referer': 'https://deepinfra.com/',
'Sec-Fetch-Dest': 'empty',
'Sec-Fetch-Mode': 'cors',
'Sec-Fetch-Site': 'same-site',
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36',
'X-Deepinfra-Source': 'web-embed',
'accept': 'text/event-stream',
'sec-ch-ua': '"Not;A=Brand";v="24", "Chromium";v="128"',
'sec-ch-ua-mobile': '?0',
'sec-ch-ua-platform': '"Linux"',
}
async with ClientSession(headers=headers) as session:
prompt = format_prompt(messages)
data = {
'model': model,
'messages': [
{'role': 'system', 'content': 'Be a helpful assistant'},
{'role': 'user', 'content': prompt}
],
'stream': True
}
if model == 'openbmb/MiniCPM-Llama3-V-2_5' and image is not None:
data['messages'][-1]['content'] = [
{
'type': 'image_url',
'image_url': {
'url': to_data_uri(image)
}
},
{
'type': 'text',
'text': messages[-1]['content']
}
]
async with session.post(cls.api_endpoint, json=data, proxy=proxy) as response:
response.raise_for_status()
async for line in response.content:
if line:
decoded_line = line.decode('utf-8').strip()
if decoded_line.startswith('data:'):
json_part = decoded_line[5:].strip()
if json_part == '[DONE]':
break
try:
data = json.loads(json_part)
choices = data.get('choices', [])
if choices:
delta = choices[0].get('delta', {})
content = delta.get('content', '')
if content:
yield content
except json.JSONDecodeError:
print(f"JSON decode error: {json_part}")
