webscout 8.3.1__py3-none-any.whl → 8.3.3__py3-none-any.whl

webscout/Provider/OPENAI/README.md

@@ -21,9 +21,9 @@
 
 The WebScout OpenAI-Compatible Providers module offers a standardized way to interact with various AI providers using the familiar OpenAI API structure. This makes it easy to:
 
-*   Use the same code structure across different AI providers
-*   Switch between providers without major code changes
-*   Leverage the OpenAI ecosystem of tools and libraries with alternative AI providers
+- Use the same code structure across different AI providers
+- Switch between providers without major code changes
+- Leverage the OpenAI ecosystem of tools and libraries with alternative AI providers
 
 ## ⚙️ Available Providers
 
@@ -68,8 +68,11 @@ Currently, the following providers are implemented with OpenAI-compatible interf
 - TogetherAI
 - PiAI
 - FalconH1
----
-
+- XenAI
+- GeminiProxy
+- MonoChat
+- Friendli
+- MiniMax
 
 ## 💻 Usage Examples
 
@@ -907,17 +910,17 @@ All providers return responses that mimic the OpenAI API structure, ensuring com
 
 The OpenAI-compatible providers are built on a modular architecture:
 
-*   `base.py`: Contains abstract base classes that define the OpenAI-compatible interface
-*   `utils.py`: Provides data structures that mimic OpenAI's response format
-*   Provider-specific implementations (e.g., `deepinfra.py`): Implement the abstract interfaces for specific providers
+- `base.py`: Contains abstract base classes that define the OpenAI-compatible interface
+- `utils.py`: Provides data structures that mimic OpenAI's response format
+- Provider-specific implementations (e.g., `deepinfra.py`): Implement the abstract interfaces for specific providers
 
 This architecture makes it easy to add new providers while maintaining a consistent interface.
 
 ## 📝 Notes
 
-*   Some providers may require API keys for full functionality
-*   Not all OpenAI features are supported by all providers
-*   Response formats are standardized to match OpenAI's format, but the underlying content depends on the specific provider and model
+- Some providers may require API keys for full functionality
+- Not all OpenAI features are supported by all providers
+- Response formats are standardized to match OpenAI's format, but the underlying content depends on the specific provider and model
 
 ## 🤝 Contributing
 
@@ -930,24 +933,24 @@ Want to add a new OpenAI-compatible provider? Follow these steps:
 
 ## 📚 Related Documentation
 
-*   [OpenAI API Reference](https://platform.openai.com/docs/api-reference)
-*   [DeepInfra Documentation](https://deepinfra.com/docs)
-*   [Glider.so Website](https://glider.so/)
-*   [ChatGPT Clone Website](https://chatgpt-clone-ten-nu.vercel.app/)
-*   [X0GPT Website](https://x0-gpt.devwtf.in/)
-*   [WiseCat Website](https://wise-cat-groq.vercel.app/)
-*   [Venice AI Website](https://venice.ai/)
-*   [ExaAI Website](https://o3minichat.exa.ai/)
-*   [TypeGPT Website](https://chat.typegpt.net/)
-*   [SciraChat Website](https://scira.ai/)
-*   [FreeAIChat Website](https://freeaichatplayground.com/)
-*   [LLMChatCo Website](https://llmchat.co/)
-*   [Yep.com Website](https://yep.com/)
-*   [HeckAI Website](https://heck.ai/)
-*   [SonusAI Website](https://chat.sonus.ai/)
-*   [ExaChat Website](https://exa-chat.vercel.app/)
-*   [Netwrck Website](https://netwrck.com/)
-*   [StandardInput Website](https://chat.standard-input.com/)
+- [OpenAI API Reference](https://platform.openai.com/docs/api-reference)
+- [DeepInfra Documentation](https://deepinfra.com/docs)
+- [Glider.so Website](https://glider.so/)
+- [ChatGPT Clone Website](https://chatgpt-clone-ten-nu.vercel.app/)
+- [X0GPT Website](https://x0-gpt.devwtf.in/)
+- [WiseCat Website](https://wise-cat-groq.vercel.app/)
+- [Venice AI Website](https://venice.ai/)
+- [ExaAI Website](https://o3minichat.exa.ai/)
+- [TypeGPT Website](https://chat.typegpt.net/)
+- [SciraChat Website](https://scira.ai/)
+- [FreeAIChat Website](https://freeaichatplayground.com/)
+- [LLMChatCo Website](https://llmchat.co/)
+- [Yep.com Website](https://yep.com/)
+- [HeckAI Website](https://heck.ai/)
+- [SonusAI Website](https://chat.sonus.ai/)
+- [ExaChat Website](https://exa-chat.vercel.app/)
+- [Netwrck Website](https://netwrck.com/)
+- [StandardInput Website](https://chat.standard-input.com/)
 
 <div align="center">
   <a href="https://t.me/PyscoutAI"><img alt="Telegram Group" src="https://img.shields.io/badge/Telegram%20Group-2CA5E0?style=for-the-badge&logo=telegram&logoColor=white"></a>

webscout/Provider/OPENAI/README_AUTOPROXY.md

@@ -0,0 +1,238 @@
+# WebScout Auto-Proxy System
+
+The WebScout Auto-Proxy system provides automatic proxy injection for all OpenAI-compatible providers. This system fetches proxies from a remote source and automatically configures them for HTTP sessions.
+
+## Features
+
+- **Automatic Proxy Injection**: All OpenAI-compatible providers automatically get proxy support
+- **Multiple HTTP Client Support**: Works with `requests`, `httpx`, and `curl_cffi`
+- **Proxy Pool Management**: Automatically fetches and caches proxies from remote source
+- **Working Proxy Detection**: Tests proxies to find working ones
+- **Easy Disable Option**: Can be disabled per provider instance or globally
+
+## How It Works
+
+The system uses a metaclass (`ProxyAutoMeta`) that automatically:
+
+1. Fetches proxies from `http://207.180.209.185:5000/ips.txt`
+2. Caches proxies for 5 minutes to avoid excessive requests
+3. Randomly selects a proxy for each provider instance
+4. Patches existing HTTP session objects with proxy configuration
+5. Provides helper methods for creating proxied sessions
+
+## Usage
+
+### Automatic Usage (Default)
+
+All OpenAI-compatible providers automatically get proxy support:
+
+```python
+from webscout.Provider.OPENAI.yep import YEPCHAT
+
+# Proxy is automatically configured
+client = YEPCHAT()
+
+# All requests will use the configured proxy
+response = client.chat.completions.create(
+    model="DeepSeek-R1-Distill-Qwen-32B",
+    messages=[{"role": "user", "content": "Hello!"}]
+)
+```
+
+### Disabling Auto-Proxy
+
+You can disable automatic proxy injection:
+
+```python
+# Disable for a specific instance
+client = YEPCHAT(disable_auto_proxy=True)
+
+# Or set a class attribute to disable for all instances
+class MyProvider(OpenAICompatibleProvider):
+    DISABLE_AUTO_PROXY = True
+```
+
+### Manual Proxy Configuration
+
+You can also provide your own proxies:
+
+```python
+custom_proxies = {
+    'http': 'http://user:pass@proxy.example.com:8080',
+    'https': 'http://user:pass@proxy.example.com:8080'
+}
+
+client = YEPCHAT(proxies=custom_proxies)
+```
+
+### Using Helper Methods
+
+Each provider instance gets helper methods for creating proxied sessions:
+
+```python
+client = YEPCHAT()
+
+# Get a requests.Session with proxies configured
+session = client.get_proxied_session()
+
+# Get a curl_cffi Session with proxies configured
+curl_session = client.get_proxied_curl_session(impersonate="chrome120")
+
+# Get an httpx.Client with proxies configured (if httpx is installed)
+httpx_client = client.get_proxied_httpx_client()
+```
+
+## Direct API Usage
+
+You can also use the proxy functions directly:
+
+```python
+from webscout.Provider.OPENAI.autoproxy import (
+    get_auto_proxy,
+    get_proxy_dict,
+    get_working_proxy,
+    test_proxy,
+    get_proxy_stats
+)
+
+# Get a random proxy
+proxy = get_auto_proxy()
+
+# Get proxy in dictionary format
+proxy_dict = get_proxy_dict()
+
+# Find a working proxy (tests multiple proxies)
+working_proxy = get_working_proxy(max_attempts=5)
+
+# Test if a proxy is working
+is_working = test_proxy(proxy)
+
+# Get proxy cache statistics
+stats = get_proxy_stats()
+```
+
+## Proxy Format
+
+The system expects proxies in the format:
+```
+http://username:password@host:port
+```
+
+Example:
+```
+http://fnXlN8NP6StpxZkxmNLyOt2MaVLQunpGC7K96j7R0KbnE5sU_2RdYRxaoy7P2yfqrD7Y8UFexv8kpTyK0LwkDQ==:fnXlN8NP6StpxZkxmNLyOt2MaVLQunpGC7K96j7R0KbnE5sU_2RdYRxaoy7P2yfqrD7Y8UFexv8kpTyK0LwkDQ==@190.103.177.163:80
+```
+
+## Configuration
+
+### Cache Duration
+
+You can adjust the proxy cache duration:
+
+```python
+from webscout.Provider.OPENAI.autoproxy import set_proxy_cache_duration
+
+# Set cache to 10 minutes
+set_proxy_cache_duration(600)
+```
+
+### Force Refresh
+
+You can force refresh the proxy cache:
+
+```python
+from webscout.Provider.OPENAI.autoproxy import refresh_proxy_cache
+
+# Force refresh and get number of proxies loaded
+count = refresh_proxy_cache()
+print(f"Loaded {count} proxies")
+```
+
+## Error Handling
+
+The system gracefully handles errors:
+
+- If proxy fetching fails, providers work without proxies
+- If a proxy test fails, the system tries other proxies
+- If no working proxy is found, providers fall back to direct connections
+
+## Logging
+
+The system uses Python's logging module. To see proxy-related logs:
+
+```python
+import logging
+logging.basicConfig(level=logging.INFO)
+
+# Or specifically for the autoproxy module
+logger = logging.getLogger('webscout.Provider.OPENAI.autoproxy')
+logger.setLevel(logging.DEBUG)
+```
+
+## Testing
+
+Run the test suite to verify functionality:
+
+```bash
+python webscout/Provider/OPENAI/test_autoproxy.py
+```
+
+## Implementation Details
+
+### ProxyAutoMeta Metaclass
+
+The `ProxyAutoMeta` metaclass is applied to `OpenAICompatibleProvider` and:
+
+1. Intercepts class instantiation
+2. Checks for `disable_auto_proxy` parameter or class attribute
+3. Fetches and configures proxies if not disabled
+4. Patches existing session objects
+5. Adds helper methods to the instance
+
+### Session Patching
+
+The system automatically patches these session types:
+- `requests.Session` - Updates the `proxies` attribute
+- `httpx.Client` - Sets the `_proxies` attribute
+- `curl_cffi.Session` - Updates the `proxies` attribute
+- `curl_cffi.AsyncSession` - Updates the `proxies` attribute
+
+### Proxy Source
+
+Proxies are fetched from: `http://207.180.209.185:5000/ips.txt`
+
+The system expects one proxy per line in the format shown above.
+
+## Troubleshooting
+
+### No Proxies Available
+
+If you see "No proxies available" messages:
+1. Check if the proxy source URL is accessible
+2. Verify your internet connection
+3. Check if the proxy format is correct
+
+### Proxy Test Failures
+
+If proxy tests fail:
+1. Some proxies may be temporarily unavailable (normal)
+2. The test URL (`https://httpbin.org/ip`) may be blocked
+3. Network connectivity issues
+
+### Provider Not Getting Proxies
+
+If a provider doesn't get automatic proxies:
+1. Ensure it inherits from `OpenAICompatibleProvider`
+2. Check if `disable_auto_proxy` is set
+3. Verify the metaclass is properly imported
+
+## Contributing
+
+To add proxy support to a new provider:
+
+1. Inherit from `OpenAICompatibleProvider`
+2. Accept `disable_auto_proxy` parameter in `__init__`
+3. Use `self.proxies` for HTTP requests
+4. Optionally use helper methods like `self.get_proxied_session()`
+
+The metaclass will handle the rest automatically!

webscout/Provider/OPENAI/TogetherAI.py

@@ -208,21 +208,15 @@ class TogetherAI(OpenAICompatibleProvider):
     OpenAI-compatible client for TogetherAI API.
     """
     AVAILABLE_MODELS = [
-        "Gryphe/MythoMax-L2-13b",
-        "Gryphe/MythoMax-L2-13b-Lite",
         "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
         "Qwen/QwQ-32B",
         "Qwen/Qwen2-72B-Instruct",
         "Qwen/Qwen2-VL-72B-Instruct",
         "Qwen/Qwen2.5-72B-Instruct-Turbo",
         "Qwen/Qwen2.5-7B-Instruct-Turbo",
-        "Qwen/Qwen2.5-Coder-32B-Instruct",
         "Qwen/Qwen2.5-VL-72B-Instruct",
-        "Qwen/Qwen3-235B-A22B-fp8",
         "Qwen/Qwen3-235B-A22B-fp8-tput",
-        "Rrrr/meta-llama/Llama-3-70b-chat-hf-6f9ad551",
-        "Rrrr/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo-03dc18e1",
-        "Rrrr/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo-6c92f39d",
+        "Salesforce/Llama-Rank-V1",
         "arcee-ai/arcee-blitz",
         "arcee-ai/caller",
         "arcee-ai/coder-large",
@@ -237,13 +231,12 @@ class TogetherAI(OpenAICompatibleProvider):
         "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
         "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
         "deepseek-ai/DeepSeek-V3",
-        "deepseek-ai/DeepSeek-V3-p-dp",
         "google/gemma-2-27b-it",
-        "google/gemma-2b-it",
         "lgai/exaone-3-5-32b-instruct",
         "lgai/exaone-deep-32b",
         "marin-community/marin-8b-instruct",
-        "meta-llama/Llama-3-70b-chat-hf",
+        "meta-llama-llama-2-70b-hf",
+        "meta-llama/Llama-2-70b-hf",
         "meta-llama/Llama-3-8b-chat-hf",
         "meta-llama/Llama-3.2-11B-Vision-Instruct-Turbo",
         "meta-llama/Llama-3.2-3B-Instruct-Turbo",
@@ -265,14 +258,8 @@ class TogetherAI(OpenAICompatibleProvider):
         "mistralai/Mixtral-8x7B-Instruct-v0.1",
         "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF",
         "perplexity-ai/r1-1776",
-        "roberizk@gmail.com/meta-llama/Llama-3-70b-chat-hf-26ee936b",
-        "roberizk@gmail.com/meta-llama/Meta-Llama-3-70B-Instruct-6feb41f7",
-        "roberizk@gmail.com/meta-llama/Meta-Llama-3-8B-Instruct-8ced8839",
         "scb10x/scb10x-llama3-1-typhoon2-70b-instruct",
-        "scb10x/scb10x-llama3-1-typhoon2-8b-instruct",
-        "togethercomputer/MoA-1",
-        "togethercomputer/MoA-1-Turbo",
-        "togethercomputer/Refuel-Llm-V2",
+        "scb10x/scb10x-typhoon-2-1-gemma3-12b",
         "togethercomputer/Refuel-Llm-V2-Small",
     ]
 

webscout/Provider/OPENAI/__init__.py

@@ -40,4 +40,20 @@ from .oivscode import * # Add OnRender provider
 from .Qwen3 import *
 from .FalconH1 import *
 from .PI import *  # Add PI.ai provider
-from .TogetherAI import *  # Add TogetherAI provider
+from .TogetherAI import *  # Add TogetherAI provider
+from .xenai import *  # Add XenAI provider
+from .GeminiProxy import *  # Add GeminiProxy provider
+from .friendli import *
+from .monochat import *
+from .MiniMax import *  # Add MiniMaxAI provider
+# Export auto-proxy functionality
+from .autoproxy import (
+    get_auto_proxy,
+    get_proxy_dict,
+    get_working_proxy,
+    test_proxy,
+    get_proxy_stats,
+    refresh_proxy_cache,
+    set_proxy_cache_duration,
+    ProxyAutoMeta
+)