gpt4free/g4f/Provider/Blackbox.py
hlohaus c97ba0c88e Add audio transcribing example and support
Add Grok Chat provider
Rename images parameter to media
Update demo homepage
2025-03-21 03:17:45 +01:00

649 lines
30 KiB
Python

from __future__ import annotations
from aiohttp import ClientSession
import os
import re
import json
import random
import string
import base64
from pathlib import Path
from typing import Optional
from datetime import datetime, timedelta
from ..typing import AsyncResult, Messages, MediaListType
from ..requests.raise_for_status import raise_for_status
from .base_provider import AsyncGeneratorProvider, ProviderModelMixin
from ..image import to_data_uri
from ..cookies import get_cookies_dir
from .helper import format_prompt, format_image_prompt
from ..providers.response import JsonConversation, ImageResponse
from ..errors import ModelNotSupportedError
from .. import debug
class Conversation(JsonConversation):
validated_value: str = None
chat_id: str = None
message_history: Messages = []
def __init__(self, model: str):
self.model = model
class Blackbox(AsyncGeneratorProvider, ProviderModelMixin):
label = "Blackbox AI"
url = "https://www.blackbox.ai"
api_endpoint = "https://www.blackbox.ai/api/chat"
working = True
supports_stream = True
supports_system_message = True
supports_message_history = True
default_model = "blackboxai"
default_vision_model = default_model
default_image_model = 'flux'
# Completely free models
fallback_models = [
"blackboxai",
"gpt-4o-mini",
"GPT-4o",
"o1",
"o3-mini",
"Claude-sonnet-3.7",
"DeepSeek-V3",
"DeepSeek-R1",
"DeepSeek-LLM-Chat-(67B)",
# Image models
"flux",
# Trending agent modes
'Python Agent',
'HTML Agent',
'Builder Agent',
'Java Agent',
'JavaScript Agent',
'React Agent',
'Android Agent',
'Flutter Agent',
'Next.js Agent',
'AngularJS Agent',
'Swift Agent',
'MongoDB Agent',
'PyTorch Agent',
'Xcode Agent',
'Azure Agent',
'Bitbucket Agent',
'DigitalOcean Agent',
'Docker Agent',
'Electron Agent',
'Erlang Agent',
'FastAPI Agent',
'Firebase Agent',
'Flask Agent',
'Git Agent',
'Gitlab Agent',
'Go Agent',
'Godot Agent',
'Google Cloud Agent',
'Heroku Agent'
]
image_models = [default_image_model]
vision_models = [default_vision_model, 'GPT-4o', 'o1', 'o3-mini', 'Gemini-PRO', 'Gemini Agent', 'llama-3.1-8b Agent', 'llama-3.1-70b Agent', 'llama-3.1-405 Agent', 'Gemini-Flash-2.0', 'DeepSeek-V3']
userSelectedModel = ['GPT-4o', 'o1', 'o3-mini', 'Gemini-PRO', 'Claude-sonnet-3.7', 'DeepSeek-V3', 'DeepSeek-R1', 'Meta-Llama-3.3-70B-Instruct-Turbo', 'Mistral-Small-24B-Instruct-2501', 'DeepSeek-LLM-Chat-(67B)', 'DBRX-Instruct', 'Qwen-QwQ-32B-Preview', 'Nous-Hermes-2-Mixtral-8x7B-DPO', 'Gemini-Flash-2.0']
# Agent mode configurations
agentMode = {
'GPT-4o': {'mode': True, 'id': "GPT-4o", 'name': "GPT-4o"},
'Gemini-PRO': {'mode': True, 'id': "Gemini-PRO", 'name': "Gemini-PRO"},
'Claude-sonnet-3.7': {'mode': True, 'id': "Claude-sonnet-3.7", 'name': "Claude-sonnet-3.7"},
'DeepSeek-V3': {'mode': True, 'id': "deepseek-chat", 'name': "DeepSeek-V3"},
'DeepSeek-R1': {'mode': True, 'id': "deepseek-reasoner", 'name': "DeepSeek-R1"},
'Meta-Llama-3.3-70B-Instruct-Turbo': {'mode': True, 'id': "meta-llama/Llama-3.3-70B-Instruct-Turbo", 'name': "Meta-Llama-3.3-70B-Instruct-Turbo"},
'Gemini-Flash-2.0': {'mode': True, 'id': "Gemini/Gemini-Flash-2.0", 'name': "Gemini-Flash-2.0"},
'Mistral-Small-24B-Instruct-2501': {'mode': True, 'id': "mistralai/Mistral-Small-24B-Instruct-2501", 'name': "Mistral-Small-24B-Instruct-2501"},
'DeepSeek-LLM-Chat-(67B)': {'mode': True, 'id': "deepseek-ai/deepseek-llm-67b-chat", 'name': "DeepSeek-LLM-Chat-(67B)"},
'DBRX-Instruct': {'mode': True, 'id': "databricks/dbrx-instruct", 'name': "DBRX-Instruct"},
'Qwen-QwQ-32B-Preview': {'mode': True, 'id': "Qwen/QwQ-32B-Preview", 'name': "Qwen-QwQ-32B-Preview"},
'Nous-Hermes-2-Mixtral-8x7B-DPO': {'mode': True, 'id': "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO", 'name': "Nous-Hermes-2-Mixtral-8x7B-DPO"},
}
# Trending agent modes
trendingAgentMode = {
"Gemini Agent": {'mode': True, 'id': 'gemini'},
"llama-3.1-405 Agent": {'mode': True, 'id': "llama-3.1-405"},
'llama-3.1-70b Agent': {'mode': True, 'id': "llama-3.1-70b"},
'llama-3.1-8b Agent': {'mode': True, 'id': "llama-3.1-8b"},
'Python Agent': {'mode': True, 'id': "python"},
'HTML Agent': {'mode': True, 'id': "html"},
'Builder Agent': {'mode': True, 'id': "builder"},
'Java Agent': {'mode': True, 'id': "java"},
'JavaScript Agent': {'mode': True, 'id': "javascript"},
'React Agent': {'mode': True, 'id': "react"},
'Android Agent': {'mode': True, 'id': "android"},
'Flutter Agent': {'mode': True, 'id': "flutter"},
'Next.js Agent': {'mode': True, 'id': "next.js"},
'AngularJS Agent': {'mode': True, 'id': "angularjs"},
'Swift Agent': {'mode': True, 'id': "swift"},
'MongoDB Agent': {'mode': True, 'id': "mongodb"},
'PyTorch Agent': {'mode': True, 'id': "pytorch"},
'Xcode Agent': {'mode': True, 'id': "xcode"},
'Azure Agent': {'mode': True, 'id': "azure"},
'Bitbucket Agent': {'mode': True, 'id': "bitbucket"},
'DigitalOcean Agent': {'mode': True, 'id': "digitalocean"},
'Docker Agent': {'mode': True, 'id': "docker"},
'Electron Agent': {'mode': True, 'id': "electron"},
'Erlang Agent': {'mode': True, 'id': "erlang"},
'FastAPI Agent': {'mode': True, 'id': "fastapi"},
'Firebase Agent': {'mode': True, 'id': "firebase"},
'Flask Agent': {'mode': True, 'id': "flask"},
'Git Agent': {'mode': True, 'id': "git"},
'Gitlab Agent': {'mode': True, 'id': "gitlab"},
'Go Agent': {'mode': True, 'id': "go"},
'Godot Agent': {'mode': True, 'id': "godot"},
'Google Cloud Agent': {'mode': True, 'id': "googlecloud"},
'Heroku Agent': {'mode': True, 'id': "heroku"},
}
# Complete list of all models (for authorized users)
_all_models = list(dict.fromkeys([
default_model,
*userSelectedModel,
*image_models,
*list(agentMode.keys()),
*list(trendingAgentMode.keys())
]))
@classmethod
def generate_session(cls, id_length: int = 21, days_ahead: int = 365) -> dict:
"""
Generate a dynamic session with proper ID and expiry format.
Args:
id_length: Length of the numeric ID (default: 21)
days_ahead: Number of days ahead for expiry (default: 365)
Returns:
dict: A session dictionary with user information and expiry
"""
# Generate numeric ID
numeric_id = ''.join(random.choice('0123456789') for _ in range(id_length))
# Generate future expiry date
future_date = datetime.now() + timedelta(days=days_ahead)
expiry = future_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
# Decode the encoded email
encoded_email = "Z2lzZWxlQGJsYWNrYm94LmFp" # Base64 encoded email
email = base64.b64decode(encoded_email).decode('utf-8')
# Generate random image ID for the new URL format
chars = string.ascii_letters + string.digits + "-"
random_img_id = ''.join(random.choice(chars) for _ in range(48))
image_url = f"https://lh3.googleusercontent.com/a/ACg8oc{random_img_id}=s96-c"
return {
"user": {
"name": "BLACKBOX AI",
"email": email,
"image": image_url,
"id": numeric_id
},
"expires": expiry
}
@classmethod
async def fetch_validated(cls, url: str = "https://www.blackbox.ai", force_refresh: bool = False) -> Optional[str]:
cache_file = Path(get_cookies_dir()) / 'blackbox.json'
if not force_refresh and cache_file.exists():
try:
with open(cache_file, 'r') as f:
data = json.load(f)
if data.get('validated_value'):
return data['validated_value']
except Exception as e:
debug.log(f"Blackbox: Error reading cache: {e}")
js_file_pattern = r'static/chunks/\d{4}-[a-fA-F0-9]+\.js'
uuid_pattern = r'["\']([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12})["\']'
def is_valid_context(text: str) -> bool:
return any(char + '=' in text for char in 'abcdefghijklmnopqrstuvwxyz')
async with ClientSession() as session:
try:
async with session.get(url) as response:
if response.status != 200:
return None
page_content = await response.text()
js_files = re.findall(js_file_pattern, page_content)
for js_file in js_files:
js_url = f"{url}/_next/{js_file}"
async with session.get(js_url) as js_response:
if js_response.status == 200:
js_content = await js_response.text()
for match in re.finditer(uuid_pattern, js_content):
start = max(0, match.start() - 10)
end = min(len(js_content), match.end() + 10)
context = js_content[start:end]
if is_valid_context(context):
validated_value = match.group(1)
cache_file.parent.mkdir(exist_ok=True)
try:
with open(cache_file, 'w') as f:
json.dump({'validated_value': validated_value}, f)
except Exception as e:
debug.log(f"Blackbox: Error writing cache: {e}")
return validated_value
except Exception as e:
debug.log(f"Blackbox: Error retrieving validated_value: {e}")
return None
@classmethod
def generate_id(cls, length: int = 7) -> str:
chars = string.ascii_letters + string.digits
return ''.join(random.choice(chars) for _ in range(length))
@classmethod
def get_models(cls) -> list:
"""
Returns a list of available models based on authorization status.
Authorized users get the full list of models.
Unauthorized users only get fallback_models.
"""
# Check if there are valid session data in HAR files
has_premium_access = cls._check_premium_access()
if has_premium_access:
# For authorized users - all models
debug.log(f"Blackbox: Returning full model list with {len(cls._all_models)} models")
return cls._all_models
else:
# For demo accounts - only free models
debug.log(f"Blackbox: Returning free model list with {len(cls.fallback_models)} models")
return cls.fallback_models
@classmethod
def _check_premium_access(cls) -> bool:
"""
Checks for an authorized session in HAR files.
Returns True if a valid session is found that differs from the demo.
"""
try:
har_dir = get_cookies_dir()
if not os.access(har_dir, os.R_OK):
return False
for root, _, files in os.walk(har_dir):
for file in files:
if file.endswith(".har"):
try:
with open(os.path.join(root, file), 'rb') as f:
har_data = json.load(f)
for entry in har_data['log']['entries']:
# Only check requests to blackbox API
if 'blackbox.ai/api' in entry['request']['url']:
if 'response' in entry and 'content' in entry['response']:
content = entry['response']['content']
if ('text' in content and
isinstance(content['text'], str) and
'"user"' in content['text'] and
'"email"' in content['text']):
try:
# Process request text
text = content['text'].strip()
if text.startswith('{') and text.endswith('}'):
text = text.replace('\\"', '"')
session_data = json.loads(text)
# Check if this is a valid session
if (isinstance(session_data, dict) and
'user' in session_data and
'email' in session_data['user']):
# Check if this is not a demo session
demo_session = cls.generate_session()
if (session_data['user'].get('email') !=
demo_session['user'].get('email')):
# This is not a demo session, so user has premium access
return True
except:
pass
except:
pass
return False
except Exception as e:
debug.log(f"Blackbox: Error checking premium access: {e}")
return False
# Initialize models with fallback_models
models = fallback_models
model_aliases = {
"gpt-4o": "GPT-4o",
"claude-3.7-sonnet": "Claude-sonnet-3.7",
"deepseek-v3": "DeepSeek-V3",
"deepseek-r1": "DeepSeek-R1",
"deepseek-chat": "DeepSeek-LLM-Chat-(67B)",
}
@classmethod
def generate_session(cls, id_length: int = 21, days_ahead: int = 365) -> dict:
"""
Generate a dynamic session with proper ID and expiry format.
Args:
id_length: Length of the numeric ID (default: 21)
days_ahead: Number of days ahead for expiry (default: 365)
Returns:
dict: A session dictionary with user information and expiry
"""
# Generate numeric ID
numeric_id = ''.join(random.choice('0123456789') for _ in range(id_length))
# Generate future expiry date
future_date = datetime.now() + timedelta(days=days_ahead)
expiry = future_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
# Decode the encoded email
encoded_email = "Z2lzZWxlQGJsYWNrYm94LmFp" # Base64 encoded email
email = base64.b64decode(encoded_email).decode('utf-8')
# Generate random image ID for the new URL format
chars = string.ascii_letters + string.digits + "-"
random_img_id = ''.join(random.choice(chars) for _ in range(48))
image_url = f"https://lh3.googleusercontent.com/a/ACg8oc{random_img_id}=s96-c"
return {
"user": {
"name": "BLACKBOX AI",
"email": email,
"image": image_url,
"id": numeric_id
},
"expires": expiry
}
@classmethod
async def fetch_validated(cls, url: str = "https://www.blackbox.ai", force_refresh: bool = False) -> Optional[str]:
cache_file = Path(get_cookies_dir()) / 'blackbox.json'
if not force_refresh and cache_file.exists():
try:
with open(cache_file, 'r') as f:
data = json.load(f)
if data.get('validated_value'):
return data['validated_value']
except Exception as e:
debug.log(f"Blackbox: Error reading cache: {e}")
js_file_pattern = r'static/chunks/\d{4}-[a-fA-F0-9]+\.js'
uuid_pattern = r'["\']([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12})["\']'
def is_valid_context(text: str) -> bool:
return any(char + '=' in text for char in 'abcdefghijklmnopqrstuvwxyz')
async with ClientSession() as session:
try:
async with session.get(url) as response:
if response.status != 200:
return None
page_content = await response.text()
js_files = re.findall(js_file_pattern, page_content)
for js_file in js_files:
js_url = f"{url}/_next/{js_file}"
async with session.get(js_url) as js_response:
if js_response.status == 200:
js_content = await js_response.text()
for match in re.finditer(uuid_pattern, js_content):
start = max(0, match.start() - 10)
end = min(len(js_content), match.end() + 10)
context = js_content[start:end]
if is_valid_context(context):
validated_value = match.group(1)
cache_file.parent.mkdir(exist_ok=True)
try:
with open(cache_file, 'w') as f:
json.dump({'validated_value': validated_value}, f)
except Exception as e:
debug.log(f"Blackbox: Error writing cache: {e}")
return validated_value
except Exception as e:
debug.log(f"Blackbox: Error retrieving validated_value: {e}")
return None
@classmethod
def generate_id(cls, length: int = 7) -> str:
chars = string.ascii_letters + string.digits
return ''.join(random.choice(chars) for _ in range(length))
@classmethod
async def create_async_generator(
cls,
model: str,
messages: Messages,
prompt: str = None,
proxy: str = None,
media: MediaListType = None,
top_p: float = None,
temperature: float = None,
max_tokens: int = None,
conversation: Conversation = None,
return_conversation: bool = False,
**kwargs
) -> AsyncResult:
model = cls.get_model(model)
headers = {
'accept': '*/*',
'accept-language': 'en-US,en;q=0.9',
'content-type': 'application/json',
'origin': 'https://www.blackbox.ai',
'referer': 'https://www.blackbox.ai/',
'user-agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36'
}
async with ClientSession(headers=headers) as session:
if conversation is None or not hasattr(conversation, "chat_id"):
conversation = Conversation(model)
conversation.validated_value = await cls.fetch_validated()
conversation.chat_id = cls.generate_id()
conversation.message_history = []
current_messages = []
for i, msg in enumerate(messages):
msg_id = conversation.chat_id if i == 0 and msg["role"] == "user" else cls.generate_id()
current_msg = {
"id": msg_id,
"content": msg["content"],
"role": msg["role"]
}
current_messages.append(current_msg)
if media is not None:
current_messages[-1]['data'] = {
"imagesData": [
{
"filePath": f"/{image_name}",
"contents": to_data_uri(image)
}
for image, image_name in media
],
"fileText": "",
"title": ""
}
# Try to get session data from HAR files
session_data = cls.generate_session() # Default fallback
session_found = False
# Look for HAR session data
har_dir = get_cookies_dir()
if os.access(har_dir, os.R_OK):
for root, _, files in os.walk(har_dir):
for file in files:
if file.endswith(".har"):
try:
with open(os.path.join(root, file), 'rb') as f:
har_data = json.load(f)
for entry in har_data['log']['entries']:
# Only look at blackbox API responses
if 'blackbox.ai/api' in entry['request']['url']:
# Look for a response that has the right structure
if 'response' in entry and 'content' in entry['response']:
content = entry['response']['content']
# Look for both regular and Google auth session formats
if ('text' in content and
isinstance(content['text'], str) and
'"user"' in content['text'] and
'"email"' in content['text'] and
'"expires"' in content['text']):
try:
# Remove any HTML or other non-JSON content
text = content['text'].strip()
if text.startswith('{') and text.endswith('}'):
# Replace escaped quotes
text = text.replace('\\"', '"')
har_session = json.loads(text)
# Check if this is a valid session object (supports both regular and Google auth)
if (isinstance(har_session, dict) and
'user' in har_session and
'email' in har_session['user'] and
'expires' in har_session):
file_path = os.path.join(root, file)
debug.log(f"Blackbox: Found session in HAR file")
session_data = har_session
session_found = True
break
except json.JSONDecodeError as e:
# Only print error for entries that truly look like session data
if ('"user"' in content['text'] and
'"email"' in content['text']):
debug.log(f"Blackbox: Error parsing likely session data: {e}")
if session_found:
break
except Exception as e:
debug.log(f"Blackbox: Error reading HAR file: {e}")
if session_found:
break
if session_found:
break
data = {
"messages": current_messages,
"agentMode": cls.agentMode.get(model, {}) if model in cls.agentMode else {},
"id": conversation.chat_id,
"previewToken": None,
"userId": None,
"codeModelMode": True,
"trendingAgentMode": cls.trendingAgentMode.get(model, {}) if model in cls.trendingAgentMode else {},
"isMicMode": False,
"userSystemPrompt": None,
"maxTokens": max_tokens,
"playgroundTopP": top_p,
"playgroundTemperature": temperature,
"isChromeExt": False,
"githubToken": "",
"clickedAnswer2": False,
"clickedAnswer3": False,
"clickedForceWebSearch": False,
"visitFromDelta": False,
"isMemoryEnabled": False,
"mobileClient": False,
"userSelectedModel": model if model in cls.userSelectedModel else None,
"validated": conversation.validated_value,
"imageGenerationMode": model == cls.default_image_model,
"webSearchModePrompt": False,
"deepSearchMode": False,
"domains": None,
"vscodeClient": False,
"codeInterpreterMode": False,
"customProfile": {
"name": "",
"occupation": "",
"traits": [],
"additionalInfo": "",
"enableNewChats": False
},
"session": session_data if session_data else cls.generate_session(),
"isPremium": True,
"subscriptionCache": None,
"beastMode": False,
"webSearchMode": False
}
# Add debugging before making the API call
if isinstance(session_data, dict) and 'user' in session_data:
# Генеруємо демо-сесію для порівняння
demo_session = cls.generate_session()
is_demo = False
if demo_session and isinstance(demo_session, dict) and 'user' in demo_session:
if session_data['user'].get('email') == demo_session['user'].get('email'):
is_demo = True
if is_demo:
debug.log(f"Blackbox: Making API request with built-in Developer Premium Account")
else:
user_email = session_data['user'].get('email', 'unknown')
debug.log(f"Blackbox: Making API request with HAR session email: {user_email}")
# Continue with the API request and async generator behavior
async with session.post(cls.api_endpoint, json=data, proxy=proxy) as response:
await raise_for_status(response)
# Collect the full response
full_response = []
async for chunk in response.content.iter_any():
if chunk:
chunk_text = chunk.decode()
full_response.append(chunk_text)
# Only yield chunks for non-image models
if model != cls.default_image_model:
yield chunk_text
full_response_text = ''.join(full_response)
# For image models, check for image markdown
if model == cls.default_image_model:
image_url_match = re.search(r'!\[.*?\]\((.*?)\)', full_response_text)
if image_url_match:
image_url = image_url_match.group(1)
yield ImageResponse(images=[image_url], alt=format_image_prompt(messages, prompt))
return
# Handle conversation history once, in one place
if return_conversation:
conversation.message_history.append({"role": "assistant", "content": full_response_text})
yield conversation
# For image models that didn't produce an image, fall back to text response
elif model == cls.default_image_model:
yield full_response_text