gpt4free/g4f/Provider/Qwen.py
2025-09-21 22:26:28 +02:00

307 lines
14 KiB
Python

from __future__ import annotations
import asyncio
import json
import re
import uuid
from time import time
from typing import Literal, Optional
import aiohttp
from ..errors import RateLimitError
from ..typing import AsyncResult, Messages, MediaListType
from ..providers.response import JsonConversation, Reasoning, Usage, ImageResponse, FinishReason
from ..requests import sse_stream
from ..tools.media import merge_media
from .base_provider import AsyncGeneratorProvider, ProviderModelMixin
from .helper import get_last_user_message
from .. import debug
try:
import curl_cffi
has_curl_cffi = True
except ImportError:
has_curl_cffi = False
text_models = [
'qwen3-max-preview', 'qwen-plus-2025-09-11', 'qwen3-235b-a22b', 'qwen3-coder-plus', 'qwen3-30b-a3b',
'qwen3-coder-30b-a3b-instruct', 'qwen-max-latest', 'qwen-plus-2025-01-25', 'qwq-32b', 'qwen-turbo-2025-02-11',
'qwen2.5-omni-7b', 'qvq-72b-preview-0310', 'qwen2.5-vl-32b-instruct', 'qwen2.5-14b-instruct-1m',
'qwen2.5-coder-32b-instruct', 'qwen2.5-72b-instruct']
image_models = [
'qwen3-max-preview', 'qwen-plus-2025-09-11', 'qwen3-235b-a22b', 'qwen3-coder-plus', 'qwen3-30b-a3b',
'qwen3-coder-30b-a3b-instruct', 'qwen-max-latest', 'qwen-plus-2025-01-25', 'qwen-turbo-2025-02-11',
'qwen2.5-omni-7b', 'qwen2.5-vl-32b-instruct', 'qwen2.5-14b-instruct-1m', 'qwen2.5-coder-32b-instruct',
'qwen2.5-72b-instruct']
vision_models = [
'qwen3-max-preview', 'qwen-plus-2025-09-11', 'qwen3-235b-a22b', 'qwen3-coder-plus', 'qwen3-30b-a3b',
'qwen3-coder-30b-a3b-instruct', 'qwen-max-latest', 'qwen-plus-2025-01-25', 'qwen-turbo-2025-02-11',
'qwen2.5-omni-7b', 'qvq-72b-preview-0310', 'qwen2.5-vl-32b-instruct', 'qwen2.5-14b-instruct-1m',
'qwen2.5-coder-32b-instruct', 'qwen2.5-72b-instruct']
models = [
'qwen3-max-preview', 'qwen-plus-2025-09-11', 'qwen3-235b-a22b', 'qwen3-coder-plus', 'qwen3-30b-a3b',
'qwen3-coder-30b-a3b-instruct', 'qwen-max-latest', 'qwen-plus-2025-01-25', 'qwq-32b', 'qwen-turbo-2025-02-11',
'qwen2.5-omni-7b', 'qvq-72b-preview-0310', 'qwen2.5-vl-32b-instruct', 'qwen2.5-14b-instruct-1m',
'qwen2.5-coder-32b-instruct', 'qwen2.5-72b-instruct']
class Qwen(AsyncGeneratorProvider, ProviderModelMixin):
"""
Provider for Qwen's chat service (chat.qwen.ai), with configurable
parameters (stream, enable_thinking) and print logs.
"""
url = "https://chat.qwen.ai"
working = True
active_by_default = True
supports_stream = True
supports_message_history = False
_models_loaded = True
image_models = image_models
text_models = text_models
vision_models = vision_models
models = models
default_model = "qwen3-235b-a22b"
_midtoken: str = None
_midtoken_uses: int = 0
@classmethod
def get_models(cls) -> list[str]:
if not cls._models_loaded and has_curl_cffi:
response = curl_cffi.get(f"{cls.url}/api/models")
if response.ok:
models = response.json().get("data", [])
cls.text_models = [model["id"] for model in models if "t2t" in model["info"]["meta"]["chat_type"]]
cls.image_models = [
model["id"] for model in models if
"image_edit" in model["info"]["meta"]["chat_type"] or "t2i" in model["info"]["meta"]["chat_type"]
]
cls.vision_models = [model["id"] for model in models if model["info"]["meta"]["capabilities"]["vision"]]
cls.models = [model["id"] for model in models]
cls.default_model = cls.models[0]
cls._models_loaded = True
cls.live += 1
debug.log(f"Loaded {len(cls.models)} models from {cls.url}")
else:
debug.log(f"Failed to load models from {cls.url}: {response.status_code} {response.reason}")
return cls.models
@classmethod
async def create_async_generator(
cls,
model: str,
messages: Messages,
media: MediaListType = None,
conversation: JsonConversation = None,
proxy: str = None,
timeout: int = 120,
stream: bool = True,
enable_thinking: bool = True,
chat_type: Literal[
"t2t", "search", "artifacts", "web_dev", "deep_research", "t2i", "image_edit", "t2v"
] = "t2t",
aspect_ratio: Optional[Literal["1:1", "4:3", "3:4", "16:9", "9:16"]] = None,
**kwargs
) -> AsyncResult:
"""
chat_type:
DeepResearch = "deep_research"
Artifacts = "artifacts"
WebSearch = "search"
ImageGeneration = "t2i"
ImageEdit = "image_edit"
VideoGeneration = "t2v"
Txt2Txt = "t2t"
WebDev = "web_dev"
"""
model_name = cls.get_model(model)
headers = {
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36',
'Accept': '*/*',
'Accept-Language': 'en-US,en;q=0.5',
'Origin': cls.url,
'Referer': f'{cls.url}/',
'Content-Type': 'application/json',
'Sec-Fetch-Dest': 'empty',
'Sec-Fetch-Mode': 'cors',
'Sec-Fetch-Site': 'same-origin',
'Connection': 'keep-alive',
'Authorization': 'Bearer',
'Source': 'web'
}
prompt = get_last_user_message(messages)
async with aiohttp.ClientSession(headers=headers) as session:
for attempt in range(5):
try:
if not cls._midtoken:
debug.log("[Qwen] INFO: No active midtoken. Fetching a new one...")
async with session.get('https://sg-wum.alibaba.com/w/wu.json', proxy=proxy) as r:
r.raise_for_status()
text = await r.text()
match = re.search(r"(?:umx\.wu|__fycb)\('([^']+)'\)", text)
if not match:
raise RuntimeError("Failed to extract bx-umidtoken.")
cls._midtoken = match.group(1)
cls._midtoken_uses = 1
debug.log(
f"[Qwen] INFO: New midtoken obtained. Use count: {cls._midtoken_uses}. Midtoken: {cls._midtoken}")
else:
cls._midtoken_uses += 1
debug.log(f"[Qwen] INFO: Reusing midtoken. Use count: {cls._midtoken_uses}")
req_headers = session.headers.copy()
req_headers['bx-umidtoken'] = cls._midtoken
req_headers['bx-v'] = '2.5.31'
message_id = str(uuid.uuid4())
if conversation is None:
chat_payload = {
"title": "New Chat",
"models": [model_name],
"chat_mode": "normal",
"chat_type": chat_type,
"timestamp": int(time() * 1000)
}
async with session.post(
f'{cls.url}/api/v2/chats/new', json=chat_payload, headers=req_headers, proxy=proxy
) as resp:
resp.raise_for_status()
data = await resp.json()
if not (data.get('success') and data['data'].get('id')):
raise RuntimeError(f"Failed to create chat: {data}")
conversation = JsonConversation(
chat_id=data['data']['id'],
cookies={key: value for key, value in resp.cookies.items()},
parent_id=None
)
files = []
media = list(merge_media(media, messages))
if media:
for _file, file_name in media:
file_class: Literal["default", "vision", "video", "audio", "document"] = "vision"
_type: Literal["file", "image", "video", "audio"] = "image"
file_type = "image/jpeg"
showType: Literal["file", "image", "video", "audio"] = "image"
if isinstance(_file, str) and _file.startswith('http'):
if chat_type == "image_edit":
file_class = "vision"
_type = "image"
file_type = "image"
showType = "image"
files.append(
{
"type": _type,
"name": file_name,
"file_type": file_type,
"showType": showType,
"file_class": file_class,
"url": _file
}
)
msg_payload = {
"stream": stream,
"incremental_output": stream,
"chat_id": conversation.chat_id,
"chat_mode": "normal",
"model": model_name,
"parent_id": conversation.parent_id,
"messages": [
{
"fid": message_id,
"parentId": conversation.parent_id,
"childrenIds": [],
"role": "user",
"content": prompt,
"user_action": "chat",
"files": files,
"models": [model_name],
"chat_type": chat_type,
"feature_config": {
"thinking_enabled": enable_thinking,
"output_schema": "phase",
"thinking_budget": 81920
},
"extra": {
"meta": {
"subChatType": chat_type
}
},
"sub_chat_type": chat_type,
"parent_id": None
}
]
}
if aspect_ratio:
msg_payload["size"] = aspect_ratio
async with session.post(
f'{cls.url}/api/v2/chat/completions?chat_id={conversation.chat_id}', json=msg_payload,
headers=req_headers, proxy=proxy, timeout=timeout, cookies=conversation.cookies
) as resp:
first_line = await resp.content.readline()
line_str = first_line.decode().strip()
if line_str.startswith('{'):
data = json.loads(line_str)
if data.get("data", {}).get("code"):
raise RuntimeError(f"Response: {data}")
conversation.parent_id = data.get("response.created", {}).get("response_id")
yield conversation
thinking_started = False
usage = None
async for chunk in sse_stream(resp):
try:
usage = chunk.get("usage", usage)
choices = chunk.get("choices", [])
if not choices: continue
delta = choices[0].get("delta", {})
phase = delta.get("phase")
content = delta.get("content")
status = delta.get("status")
extra = delta.get("extra", {})
if phase == "think" and not thinking_started:
thinking_started = True
elif phase == "answer" and thinking_started:
thinking_started = False
elif phase == "image_gen" and status == "typing":
yield ImageResponse(content, prompt, extra)
continue
elif phase == "image_gen" and status == "finished":
yield FinishReason("stop")
if content:
yield Reasoning(content) if thinking_started else content
except (json.JSONDecodeError, KeyError, IndexError):
continue
if usage:
yield Usage(**usage)
return
except (aiohttp.ClientResponseError, RuntimeError) as e:
is_rate_limit = (isinstance(e, aiohttp.ClientResponseError) and e.status == 429) or \
("RateLimited" in str(e))
if is_rate_limit:
debug.log(
f"[Qwen] WARNING: Rate limit detected (attempt {attempt + 1}/5). Invalidating current midtoken.")
cls._midtoken = None
cls._midtoken_uses = 0
conversation = None
await asyncio.sleep(2)
continue
else:
raise e
raise RateLimitError("The Qwen provider reached the request limit after 5 attempts.")