llm-onesdk 0.0.1

OneSDK is a Python library that provides a unified interface for interacting with various Large Language Model (LLM) providers.

OneSDK: Unified LLM API Interface

OneSDK is a Python library that provides a unified interface for interacting with various Large Language Model (LLM) providers. It simplifies the process of working with different LLM APIs by offering a consistent set of methods across providers.

Features

• Unified API for multiple LLM providers
• Flexible usage patterns: specify model per call or set a default model
• Easy-to-use interface for common LLM operations
• Support for both synchronous and streaming text generation
• Token counting functionality
• File upload capability (where supported)
• Proxy setting for API calls
• Usage statistics retrieval (where available)

Installation

pip install llm-onesdk

Quick Start

OneSDK supports two main usage patterns:

Pattern 1: Specify model for each call

from llm-onesdk import OneSDK

# Initialize the SDK with your chosen provider and credentials
sdk = OneSDK("anthropic", {"api_key": "your-api-key"})

# Generate text
response = sdk.generate(
    model="claude-2",
    messages=[{"role": "user", "content": "Tell me a joke about programming."}]
)
print(response['choices'][0]['message']['content'])

Pattern 2: Set a default model

from llm-onesdk import OneSDK

# Initialize the SDK and set a default model
sdk = OneSDK("anthropic", {"api_key": "your-api-key"})
sdk.set_model("claude-2")

# Generate text using the default model
response = sdk.generate(
    messages=[{"role": "user", "content": "Tell me a joke about programming."}]
)
print(response['choices'][0]['message']['content'])

Streaming Generation

Both patterns support streaming generation:

for chunk in sdk.stream_generate(
    model="claude-2",  # or omit if using set_model()
    messages=[{"role": "user", "content": "Write a short story about AI."}]
):
    print(chunk['choices'][0]['message']['content'], end='', flush=True)

Other Operations

# List available models
models = sdk.list_models()
print(models)

# Count tokens
token_count = sdk.count_tokens(
    model="claude-2",
    messages=[{"role": "user", "content": "How many tokens is this?"}]
)
print(f"Token count: {token_count}")

Supported Providers

You can list all available providers using:

print(OneSDK.list_providers())

Methods

• set_model(model): Set the default model for subsequent API calls.
• list_models(): List available models for the current provider.
• get_model_info(model_id): Get information about a specific model.
• generate(messages, model=None, **kwargs): Generate a response using the specified or default model.
• stream_generate(messages, model=None, **kwargs): Generate a streaming response.
• count_tokens(model, messages): Count the number of tokens in the input messages.
• create_completion(model, prompt, **kwargs): Create a completion using the legacy API (if supported).
• upload_file(file_path): Upload a file and return a reference.
• set_proxy(proxy_url): Set a proxy for API calls.
• get_usage(): Get usage statistics for the current account.

Error Handling

OneSDK uses custom exception classes for error handling. The base exception class is InvokeError, with specific subclasses for different error types (e.g., InvokeModelNotFoundError).

Contributing

We welcome contributions, especially new provider integrations! Please see our Contributing Guide for more information on how to get started.

License

This project is licensed under the MIT License - see the LICENSE file for details.