indoxrouter 0.1.26__py3-none-any.whl → 0.1.28__py3-none-any.whl

indoxrouter/__init__.py

@@ -47,7 +47,7 @@ from .exceptions import (
     APIError,
 )
 
-__version__ = "0.1.26"
+__version__ = "0.1.27"
 __all__ = [
     "Client",
     "IndoxRouter",

indoxrouter/client.py

@@ -14,6 +14,7 @@ The Client class offers methods for:
 - Accessing AI capabilities: chat completions, text completions, embeddings, image generation, and text-to-speech
 - Retrieving information about available providers and models
 - Monitoring usage statistics and credit consumption
+- BYOK (Bring Your Own Key) support for using your own provider API keys
 
 Usage example:
     ```python
@@ -37,6 +38,14 @@ Usage example:
     # Generate text-to-speech audio
     audio = client.text_to_speech("Hello, welcome to IndoxRouter!", model="openai/tts-1", voice="alloy")
 
+    # Transcribe audio to text using speech-to-text
+    transcription = client.speech_to_text("path/to/audio.mp3", model="openai/whisper-1")
+
+    # Using BYOK (Bring Your Own Key)
+    response = client.chat([
+        {"role": "user", "content": "Hello!"}
+    ], model="openai/gpt-4", byok_api_key="sk-your-openai-key-here")
+
     # Clean up resources when done
     client.close()
     ```
@@ -46,6 +55,21 @@ The client can also be used as a context manager:
     with Client(api_key="your_api_key") as client:
         response = client.chat([{"role": "user", "content": "Hello!"}], model="openai/gpt-4o-mini")
     ```
+
+BYOK (Bring Your Own Key) Support:
+    The client supports BYOK, allowing you to use your own API keys for AI providers:
+
+    - No credit deduction from your IndoxRouter account
+    - No rate limiting from the platform
+    - Direct provider access with your own API keys
+    - Cost control - you pay providers directly at their rates
+
+    Example:
+        response = client.chat(
+            messages=[{"role": "user", "content": "Hello!"}],
+            model="openai/gpt-4",
+            byok_api_key="sk-your-openai-key-here"
+        )
 """
 
 import os
@@ -76,11 +100,14 @@ from .constants import (
     DEFAULT_EMBEDDING_MODEL,
     DEFAULT_IMAGE_MODEL,
     DEFAULT_TTS_MODEL,
+    DEFAULT_STT_MODEL,
     CHAT_ENDPOINT,
     COMPLETION_ENDPOINT,
     EMBEDDING_ENDPOINT,
     IMAGE_ENDPOINT,
     TTS_ENDPOINT,
+    STT_ENDPOINT,
+    STT_TRANSLATION_ENDPOINT,
     MODEL_ENDPOINT,
     USAGE_ENDPOINT,
     USE_COOKIES,
@@ -236,6 +263,7 @@ class Client:
         endpoint: str,
         data: Optional[Dict[str, Any]] = None,
         stream: bool = False,
+        files: Optional[Dict[str, Any]] = None,
     ) -> Any:
         """
         Make a request to the API.
@@ -245,6 +273,7 @@ class Client:
             endpoint: API endpoint
             data: Request data
             stream: Whether to stream the response
+            files: Files to upload (for multipart/form-data requests)
 
         Returns:
             Response data
@@ -258,7 +287,14 @@ class Client:
             endpoint = endpoint[1:]
 
         url = f"{self.base_url}/{endpoint}"
-        headers = {"Content-Type": "application/json"}
+
+        # Set headers based on whether we're uploading files
+        if files:
+            # For multipart/form-data, don't set Content-Type header
+            # requests will set it automatically with boundary
+            headers = {}
+        else:
+            headers = {"Content-Type": "application/json"}
 
         # Add Authorization header if we have an access token
         if hasattr(self, "access_token") and self.access_token:
@@ -268,8 +304,8 @@ class Client:
         # if data:
         #     logger.debug(f"Request data: {json.dumps(data, indent=2)}")
 
-        # Diagnose potential issues with the request
-        if method == "POST" and data:
+        # Diagnose potential issues with the request (only for non-file uploads)
+        if method == "POST" and data and not files:
             diagnosis = self.diagnose_request(endpoint, data)
             if not diagnosis["is_valid"]:
                 issues_str = "\n".join([f"- {issue}" for issue in diagnosis["issues"]])
@@ -277,14 +313,25 @@ class Client:
                 # We'll still send the request, but log the issues
 
         try:
-            response = self.session.request(
-                method,
-                url,
-                headers=headers,
-                json=data,
-                timeout=self.timeout,
-                stream=stream,
-            )
+            # Prepare request parameters
+            request_params = {
+                "method": method,
+                "url": url,
+                "headers": headers,
+                "timeout": self.timeout,
+                "stream": stream,
+            }
+
+            # Add data based on request type
+            if files:
+                # For file uploads, use form data
+                request_params["data"] = data
+                request_params["files"] = files
+            else:
+                # For regular requests, use JSON
+                request_params["json"] = data
+
+            response = self.session.request(**request_params)
 
             if stream:
                 return response
@@ -297,16 +344,10 @@ class Client:
                 # Update Authorization header with new token if available
                 if hasattr(self, "access_token") and self.access_token:
                     headers["Authorization"] = f"Bearer {self.access_token}"
+                    request_params["headers"] = headers
 
                 # Retry the request after reauthentication
-                response = self.session.request(
-                    method,
-                    url,
-                    headers=headers,
-                    json=data,
-                    timeout=self.timeout,
-                    stream=stream,
-                )
+                response = self.session.request(**request_params)
 
                 if stream:
                     return response
@@ -459,6 +500,7 @@ class Client:
         temperature: float = 0.7,
         max_tokens: Optional[int] = None,
         stream: bool = False,
+        byok_api_key: Optional[str] = None,
         **kwargs,
     ) -> Dict[str, Any]:
         """
@@ -470,6 +512,7 @@ class Client:
             temperature: Sampling temperature
             max_tokens: Maximum number of tokens to generate
             stream: Whether to stream the response
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -490,6 +533,7 @@ class Client:
             "temperature": temperature,
             "max_tokens": max_tokens,
             "stream": stream,
+            "byok_api_key": byok_api_key,
             "additional_params": filtered_kwargs,
         }
 
@@ -506,6 +550,7 @@ class Client:
         temperature: float = 0.7,
         max_tokens: Optional[int] = None,
         stream: bool = False,
+        byok_api_key: Optional[str] = None,
         **kwargs,
     ) -> Dict[str, Any]:
         """
@@ -517,6 +562,7 @@ class Client:
             temperature: Sampling temperature
             max_tokens: Maximum number of tokens to generate
             stream: Whether to stream the response
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -537,6 +583,7 @@ class Client:
             "temperature": temperature,
             "max_tokens": max_tokens,
             "stream": stream,
+            "byok_api_key": byok_api_key,
             "additional_params": filtered_kwargs,
         }
 
@@ -550,6 +597,7 @@ class Client:
         self,
         text: Union[str, List[str]],
         model: str = DEFAULT_EMBEDDING_MODEL,
+        byok_api_key: Optional[str] = None,
         **kwargs,
     ) -> Dict[str, Any]:
         """
@@ -558,6 +606,7 @@ class Client:
         Args:
             text: Text to embed (string or list of strings)
             model: Model to use in the format "provider/model" (e.g., "openai/text-embedding-ada-002")
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -575,6 +624,7 @@ class Client:
         data = {
             "text": text if isinstance(text, list) else [text],
             "model": formatted_model,
+            "byok_api_key": byok_api_key,
             "additional_params": filtered_kwargs,
         }
 
@@ -610,6 +660,7 @@ class Client:
         enhance_prompt: Optional[bool] = None,
         # Google-specific direct parameters
         aspect_ratio: Optional[str] = None,
+        byok_api_key: Optional[str] = None,
         **kwargs,
     ) -> Dict[str, Any]:
         """
@@ -650,6 +701,8 @@ class Client:
             enhance_prompt: Whether to use prompt rewriting logic
             aspect_ratio: Aspect ratio for Google models (e.g., "1:1", "16:9") - preferred over size
 
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
+
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -674,6 +727,7 @@ class Client:
         data = {
             "prompt": prompt,
             "model": formatted_model,
+            "byok_api_key": byok_api_key,
         }
 
         # Add optional parameters only if they are explicitly provided
@@ -793,6 +847,7 @@ class Client:
         response_format: Optional[str] = None,
         speed: Optional[float] = None,
         instructions: Optional[str] = None,
+        byok_api_key: Optional[str] = None,
         **kwargs,
     ) -> Dict[str, Any]:
         """
@@ -805,6 +860,7 @@ class Client:
             response_format: Format of the audio response (e.g., "mp3", "opus", "aac", "flac")
             speed: Speed of the generated audio (0.25 to 4.0)
             instructions: Optional instructions for the TTS generation
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -829,6 +885,13 @@ class Client:
                     model="provider/model-name",
                     voice="provider-specific-voice"
                 )
+
+            Using BYOK (Bring Your Own Key):
+                response = client.text_to_speech(
+                    "Hello, world!",
+                    model="openai/tts-1",
+                    byok_api_key="sk-your-openai-key-here"
+                )
         """
         # Format the model string
         formatted_model = self._format_model_string(model)
@@ -859,8 +922,237 @@ class Client:
         if filtered_kwargs:
             data["additional_params"] = filtered_kwargs
 
+        # Add BYOK API key if provided
+        if byok_api_key:
+            data["byok_api_key"] = byok_api_key
+
         return self._request("POST", TTS_ENDPOINT, data)
 
+    def speech_to_text(
+        self,
+        file: Union[str, bytes],
+        model: str = DEFAULT_STT_MODEL,
+        language: Optional[str] = None,
+        prompt: Optional[str] = None,
+        response_format: Optional[str] = "json",
+        temperature: Optional[float] = 0.0,
+        timestamp_granularities: Optional[List[str]] = None,
+        byok_api_key: Optional[str] = None,
+        **kwargs,
+    ) -> Dict[str, Any]:
+        """
+        Transcribe audio to text using speech-to-text models.
+
+        Args:
+            file: Audio file path (str) or audio file data (bytes)
+            model: Model to use in the format "provider/model" (e.g., "openai/whisper-1")
+            language: Language code for the audio (e.g., "en", "es", "fr")
+            prompt: Optional text to guide the model's style
+            response_format: Format of the response ("json", "text", "srt", "verbose_json", "vtt")
+            temperature: Temperature for transcription (0.0 to 1.0)
+            timestamp_granularities: List of timestamp granularities (["word", "segment"])
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
+            **kwargs: Additional parameters to pass to the API
+
+        Returns:
+            Response data with transcription text
+
+        Examples:
+            Basic usage with file path:
+                response = client.speech_to_text("path/to/audio.mp3")
+
+            Basic usage with file bytes:
+                with open("audio.mp3", "rb") as f:
+                    audio_data = f.read()
+                response = client.speech_to_text(audio_data)
+
+            With specific model and language:
+                response = client.speech_to_text(
+                    "path/to/audio.wav",
+                    model="openai/whisper-1",
+                    language="en",
+                    response_format="json"
+                )
+
+            With timestamps for detailed analysis:
+                response = client.speech_to_text(
+                    "path/to/audio.mp3",
+                    model="openai/whisper-1",
+                    response_format="verbose_json",
+                    timestamp_granularities=["word", "segment"]
+                )
+
+            Using BYOK (Bring Your Own Key):
+                response = client.speech_to_text(
+                    "path/to/audio.mp3",
+                    model="openai/whisper-1",
+                    byok_api_key="sk-your-openai-key-here"
+                )
+        """
+        # Format the model string
+        formatted_model = self._format_model_string(model)
+
+        # Handle file input - can be a file path (str) or file data (bytes)
+        if isinstance(file, str):
+            # It's a file path, read the file
+            try:
+                with open(file, "rb") as f:
+                    file_data = f.read()
+                filename = os.path.basename(file)
+            except FileNotFoundError:
+                raise InvalidParametersError(f"File not found: {file}")
+            except Exception as e:
+                raise InvalidParametersError(f"Error reading file {file}: {str(e)}")
+        elif isinstance(file, bytes):
+            # It's file data
+            file_data = file
+            filename = kwargs.get("filename", "audio_file")
+        else:
+            raise InvalidParametersError(
+                "File must be either a file path (str) or file data (bytes)"
+            )
+
+        # Prepare form data for multipart upload
+        files = {"file": (filename, file_data, "audio/*")}
+
+        # Create the form data with required parameters
+        data = {
+            "model": formatted_model,
+        }
+
+        # Add optional parameters only if they are provided
+        if language is not None:
+            data["language"] = language
+        if prompt is not None:
+            data["prompt"] = prompt
+        if response_format is not None:
+            data["response_format"] = response_format
+        if temperature is not None:
+            data["temperature"] = temperature
+        if timestamp_granularities is not None:
+            # Convert to JSON string as expected by the API
+            data["timestamp_granularities"] = json.dumps(timestamp_granularities)
+
+        # Add BYOK API key if provided
+        if byok_api_key:
+            data["byok_api_key"] = byok_api_key
+
+        # Filter out problematic parameters from kwargs
+        filtered_kwargs = {}
+        for key, value in kwargs.items():
+            if key not in [
+                "filename",
+                "return_generator",
+            ]:  # List of parameters to exclude
+                filtered_kwargs[key] = value
+
+        # Add any additional parameters from kwargs
+        if filtered_kwargs:
+            data.update(filtered_kwargs)
+
+        return self._request("POST", STT_ENDPOINT, data, files=files)
+
+    def translate_audio(
+        self,
+        file: Union[str, bytes],
+        model: str = DEFAULT_STT_MODEL,
+        prompt: Optional[str] = None,
+        response_format: Optional[str] = "json",
+        temperature: Optional[float] = 0.0,
+        byok_api_key: Optional[str] = None,
+        **kwargs,
+    ) -> Dict[str, Any]:
+        """
+        Translate audio to English text using speech-to-text models.
+
+        Args:
+            file: Audio file path (str) or audio file data (bytes)
+            model: Model to use in the format "provider/model" (e.g., "openai/whisper-1")
+            prompt: Optional text to guide the model's style
+            response_format: Format of the response ("json", "text", "srt", "verbose_json", "vtt")
+            temperature: Temperature for translation (0.0 to 1.0)
+            byok_api_key: Your own API key for the provider (BYOK - Bring Your Own Key)
+            **kwargs: Additional parameters to pass to the API
+
+        Returns:
+            Response data with translated text in English
+
+        Examples:
+            Basic usage with file path:
+                response = client.translate_audio("path/to/spanish_audio.mp3")
+
+            With specific response format:
+                response = client.translate_audio(
+                    "path/to/french_audio.wav",
+                    model="openai/whisper-1",
+                    response_format="text"
+                )
+
+            Using BYOK (Bring Your Own Key):
+                response = client.translate_audio(
+                    "path/to/audio.mp3",
+                    model="openai/whisper-1",
+                    byok_api_key="sk-your-openai-key-here"
+                )
+        """
+        # Format the model string
+        formatted_model = self._format_model_string(model)
+
+        # Handle file input - can be a file path (str) or file data (bytes)
+        if isinstance(file, str):
+            # It's a file path, read the file
+            try:
+                with open(file, "rb") as f:
+                    file_data = f.read()
+                filename = os.path.basename(file)
+            except FileNotFoundError:
+                raise InvalidParametersError(f"File not found: {file}")
+            except Exception as e:
+                raise InvalidParametersError(f"Error reading file {file}: {str(e)}")
+        elif isinstance(file, bytes):
+            # It's file data
+            file_data = file
+            filename = kwargs.get("filename", "audio_file")
+        else:
+            raise InvalidParametersError(
+                "File must be either a file path (str) or file data (bytes)"
+            )
+
+        # Prepare form data for multipart upload
+        files = {"file": (filename, file_data, "audio/*")}
+
+        # Create the form data with required parameters
+        data = {
+            "model": formatted_model,
+        }
+
+        # Add optional parameters only if they are provided
+        if prompt is not None:
+            data["prompt"] = prompt
+        if response_format is not None:
+            data["response_format"] = response_format
+        if temperature is not None:
+            data["temperature"] = temperature
+
+        # Add BYOK API key if provided
+        if byok_api_key:
+            data["byok_api_key"] = byok_api_key
+
+        # Filter out problematic parameters from kwargs
+        filtered_kwargs = {}
+        for key, value in kwargs.items():
+            if key not in [
+                "filename",
+                "return_generator",
+            ]:  # List of parameters to exclude
+                filtered_kwargs[key] = value
+
+        # Add any additional parameters from kwargs
+        if filtered_kwargs:
+            data.update(filtered_kwargs)
+
+        return self._request("POST", STT_TRANSLATION_ENDPOINT, data, files=files)
+
     def _get_supported_parameters_for_model(
         self, provider: str, model_name: str
     ) -> List[str]:
@@ -875,7 +1167,6 @@ class Client:
         Returns:
             List of parameter names supported by the model
         """
-        # Define supported parameters for specific models
         if provider.lower() == "openai" and "gpt-image" in model_name.lower():
             return [
                 "prompt",

indoxrouter/constants.py

@@ -5,7 +5,7 @@ Constants for the IndoxRouter client.
 # API settings
 DEFAULT_API_VERSION = "v1"
 DEFAULT_BASE_URL = "https://api.indoxrouter.com"  # Production server URL with HTTPS
-# DEFAULT_BASE_URL = "http://localhost:8000"  # Local development server
+# DEFAULT_BASE_URL = "http://localhost:9050"  # Local development server
 DEFAULT_TIMEOUT = 60
 USE_COOKIES = True  # Always use cookie-based authentication
 
@@ -14,6 +14,7 @@ DEFAULT_MODEL = "openai/gpt-4o-mini"
 DEFAULT_EMBEDDING_MODEL = "openai/text-embedding-3-small"
 DEFAULT_IMAGE_MODEL = "openai/dall-e-3"
 DEFAULT_TTS_MODEL = "openai/tts-1"
+DEFAULT_STT_MODEL = "openai/whisper-1"
 GOOGLE_IMAGE_MODEL = "google/imagen-3.0-generate-002"
 XAI_IMAGE_MODEL = "xai/grok-2-image"
 XAI_IMAGE_LATEST_MODEL = "xai/grok-2-image-latest"
@@ -25,6 +26,8 @@ COMPLETION_ENDPOINT = "completions"
 EMBEDDING_ENDPOINT = "embeddings"
 IMAGE_ENDPOINT = "images/generations"
 TTS_ENDPOINT = "tts/generations"
+STT_ENDPOINT = "audio/stt/transcriptions"
+STT_TRANSLATION_ENDPOINT = "audio/stt/translations"
 MODEL_ENDPOINT = "models"
 USAGE_ENDPOINT = "user/usage"
 

{indoxrouter-0.1.26.dist-info → indoxrouter-0.1.28.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: indoxrouter
-Version: 0.1.26
+Version: 0.1.28
 Summary: A unified client for various AI providers
 Author-email: indoxRouter Team <ashkan.eskandari.dev@gmail.com>
 License: MIT
@@ -33,6 +33,7 @@ A unified client for various AI providers, including OpenAI, anthropic, Google,
 - **Simple Interface**: Easy-to-use methods for chat, completion, embeddings, image generation, and text-to-speech
 - **Error Handling**: Standardized error handling across providers
 - **Authentication**: Secure cookie-based authentication
+- **BYOK Support**: Bring Your Own Key support for using your own provider API keys
 
 ## Installation
 
@@ -131,6 +132,34 @@ if "b64_json" in response["data"][0]:
     # Use the base64 data (e.g., to display in HTML or save to file)
 ```
 
+### BYOK (Bring Your Own Key) Support
+
+IndoxRouter supports BYOK, allowing you to use your own API keys for AI providers:
+
+```python
+# Use your own OpenAI API key
+response = client.chat(
+    messages=[{"role": "user", "content": "Hello!"}],
+    model="openai/gpt-4",
+    byok_api_key="sk-your-openai-key-here"
+)
+
+# Use your own Google API key for image generation
+response = client.images(
+    prompt="A beautiful sunset",
+    model="google/imagen-3.0-generate-002",
+    aspect_ratio="16:9",
+    byok_api_key="your-google-api-key-here"
+)
+```
+
+**BYOK Benefits:**
+
+- No credit deduction from your IndoxRouter account
+- No platform rate limiting
+- Direct provider access with your own API keys
+- Cost control - pay providers directly at their rates
+
 ### Text-to-Speech
 
 ```python

indoxrouter-0.1.28.dist-info/RECORD

@@ -0,0 +1,9 @@
+indoxrouter/__init__.py,sha256=P_2eiiAi-DrkkeeJndQdEMqnVheWzOZIjWkDayrwxuk,1561
+indoxrouter/client.py,sha256=dzeLN-DKHiHORhXA0zI_fFg5_Rvrd98uNXqpIFKobBQ,55234
+indoxrouter/constants.py,sha256=Ti-XzQRgFQlNkfmaCKtfBhReb5nIkdnfuTvH0Zu7GHI,1584
+indoxrouter/exceptions.py,sha256=cGlXNF8dd4X8UBWgxTA099nRhEkIjcqE_4iGOlIVVp8,1632
+indoxrouter-0.1.28.dist-info/licenses/LICENSE,sha256=5n28CfoynFakg-QJIHnecEXcveN8gq-ZwhC0h7ATse0,24232
+indoxrouter-0.1.28.dist-info/METADATA,sha256=1FhWrzOdVaSLO6H29KYJBLG5kO_MQqbutpkhvh669Vc,5740
+indoxrouter-0.1.28.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+indoxrouter-0.1.28.dist-info/top_level.txt,sha256=v6FGWkw0QAnXhyYtnXLI1cxzna0iveNvZUotVzCWabM,12
+indoxrouter-0.1.28.dist-info/RECORD,,

indoxrouter-0.1.26.dist-info/RECORD

@@ -1,9 +0,0 @@
-indoxrouter/__init__.py,sha256=NhFlmJxbxcM3XCF21YZv9QsOk-k50rsqzw5Bk3mmr6E,1561
-indoxrouter/client.py,sha256=KQpOFaTntNYmPgYjBJHq0z921vgKlyeUcaTYgZX7axM,43764
-indoxrouter/constants.py,sha256=HqTXfNpL_UvWsR5mod8YZgWOs1W6x6itr1BSbmyYEHY,1451
-indoxrouter/exceptions.py,sha256=cGlXNF8dd4X8UBWgxTA099nRhEkIjcqE_4iGOlIVVp8,1632
-indoxrouter-0.1.26.dist-info/licenses/LICENSE,sha256=5n28CfoynFakg-QJIHnecEXcveN8gq-ZwhC0h7ATse0,24232
-indoxrouter-0.1.26.dist-info/METADATA,sha256=nzfcnbrzEJznKEPA1gAG34-C5KjemPqUavXO2e8-3fs,4909
-indoxrouter-0.1.26.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-indoxrouter-0.1.26.dist-info/top_level.txt,sha256=v6FGWkw0QAnXhyYtnXLI1cxzna0iveNvZUotVzCWabM,12
-indoxrouter-0.1.26.dist-info/RECORD,,