ollama/ollama-python
Ollama Python
ollama-python is a Python client for interacting with large language models. It provides an interface for sending prompts to receive text and chat completions, as well as a dedicated client for generating numerical vector embeddings from text.
The project includes a wrapper that emulates the OpenAI API, allowing applications built for that standard to interact with local models. It also provides a non-blocking asynchronous client for executing concurrent requests.
The library covers the full model lifecycle, including the ability to pull, create, list, and delete models within a local environment. It handles remote endpoint connectivity via configurable host addresses and authentication headers.
Features
- Local Model Integrations - Provides a Python client for integrating locally hosted language models to generate text and chat completions.
- OpenAI-Compatible APIs - Implements OpenAI-compatible API interfaces to allow local models to work with applications built for OpenAI.
- Chat Completion Services - Provides interfaces for sending prompts to receive both structured text responses and conversational chat replies.
- Embedding Clients - Provides a dedicated client for generating numerical vector embeddings from text using hosted models.
- Local Model Lifecycle Managers - Automates the lifecycle of local models, including pulling, creating, listing, and deleting instances.
- Model Lifecycle Managers - Implements tools for pulling, creating, listing, and deleting language models within a local environment.
- Text Embedding Generators - Includes a dedicated client for generating numerical vector embeddings from text for search and similarity tasks.
- Vector Embeddings - Converts text into numerical vector representations for use in search systems or similarity-based applications.
- LLM Clients - Provides a programmatic interface for sending prompts to large language models and receiving completions.
- Cloud API Emulators - Maps internal method calls to external API standards to ensure compatibility with third-party provider formats.
- LLM Stream Processing - Processes model responses piece by piece using Python generators for real-time text streaming.
- Unified Model Catalogs - Allows retrieval of a detailed catalog of installed models, including names, sizes, and version digests.
- Asynchronous Clients - Offers a non-blocking interface for executing concurrent requests to a language model server.
- Asynchronous Request Execution - Provides a non-blocking asynchronous client for executing concurrent requests to language models.
- Concurrent AI Requesting - Enables non-blocking requests to language models to handle multiple concurrent AI operations.
- Asyncio Coroutine Execution - Uses Python's asyncio event loop to handle multiple network requests simultaneously without halting the program.
- REST Clients - Implements an HTTP-based REST client to communicate with the backend Ollama server.
