semantic-kernel
Semantic Kernel Python SDK
Decision gist · record as of 2026-08-14
Yes. Semantic Kernel is production-stable (Development Status 5), actively maintained, MIT-licensed, and widely used (top 5000 PyPI packages). Install it if you're building AI agents, need multi-provider LLM support, or want a structured framework for prompt orchestration. The 22 runtime dependencies are substantial but standard for async AI workloads; verify your environment can accommodate them. Note that Microsoft now recommends Microsoft Agent Framework as the forward path for new enterprise projects.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10+.
- Needs API keys for LLM services (OpenAI, Azure OpenAI, etc.) set as environment variables or passed to service constructors.
- Low friction install with a pure-Python wheel.
License · maintenance · safety
permissive license (permissive) — MIT license (permissive). No restrictions on commercial or proprietary use.
last release 2026-08-06 (8 days) · last repo commit 2026-08-11 · 28,446 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 2,765,519 downloads/mo, #2,897 on PyPI
Alternatives
Verify before relying
pip install semantic-kernel
import asyncio
from semantic_kernel import Kernel
from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion
kernel = Kernel()
kernel.add_service(OpenAIChatCompletion())
result = await kernel.invoke_prompt("Hello, world!")- Performance characteristics and latency profile for multi-agent orchestration at scale.
- Compatibility matrix with specific LLM provider versions and model families beyond those listed.
- Cost implications of using vector DB integrations (Azure AI Search, Elasticsearch, Chroma) with this framework.
What it is and what it does
Semantic Kernel is Microsoft's Python SDK for building and orchestrating AI agents powered by large language models. It abstracts away the complexity of connecting to multiple LLM providers (OpenAI, Azure OpenAI, Hugging Face, Mistral, Google AI, ONNX, Ollama, NVIDIA NIM) and provides a unified interface for prompt engineering, function calling, and structured outputs through Pydantic models.
The framework enables you to compose agents with plugins (Python functions or OpenAPI specs), coordinate multi-agent workflows through orchestration managers, and build structured business processes. It handles async execution natively, integrates with vector databases for retrieval-augmented generation, and supports multimodal inputs (text, vision, audio). The package is actively maintained, recently transitioned to Microsoft Agent Framework as its enterprise successor, and comes with extensive examples for single-agent and multi-agent scenarios.
Use it for
- Build a chatbot that invokes custom Python functions or external APIs as tools in response to user queries.
- Orchestrate multiple specialized AI agents (writer, reviewer, critic) to iteratively refine content or solve problems collaboratively.
- Create a prompt engineering pipeline that templates and invokes LLM calls with variable substitution and structured response parsing.
- Integrate retrieval-augmented generation by connecting agents to vector databases for knowledge-grounded question answering.
- Deploy multi-provider LLM applications that can switch between OpenAI, Azure OpenAI, or open-source models without code changes.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
Semantic Kernel is production-stable (Development Status 5), actively maintained, MIT-licensed, and widely used (top 5000 PyPI packages). Install it if you're building AI agents, need multi-provider LLM support, or want a structured framework for prompt orchestration. The 22 runtime dependencies are substantial but standard for async AI workloads; verify your environment can accommodate them. Note that Microsoft now recommends Microsoft Agent Framework as the forward path for new enterprise projects.
Install
semantic-kernel on PyPI
Before you install
Low friction install with a pure-Python wheel. Active maintenance—released 8 days ago with 28446 GitHub stars. Requires Python 3.10+ and brings 22 runtime dependencies including Azure AI, OpenAI, Pydantic, and async tooling (aiohttp, websockets, aiortc).
Requires Python 3.10+. Needs API keys for LLM services (OpenAI, Azure OpenAI, etc.) set as environment variables or passed to service constructors.
License in practice
MIT license (permissive). No restrictions on commercial or proprietary use.
Quickstart
pip install semantic-kernel
import asyncio
from semantic_kernel import Kernel
from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion
kernel = Kernel()
kernel.add_service(OpenAIChatCompletion())
result = await kernel.invoke_prompt("Hello, world!")Verify before relying
- Performance characteristics and latency profile for multi-agent orchestration at scale.
- Compatibility matrix with specific LLM provider versions and model families beyond those listed.
- Cost implications of using vector DB integrations (Azure AI Search, Elasticsearch, Chroma) with this framework.
Package facts
| License | permissive license permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 22 packagesazure-ai-projectsazure-ai-agentsaiohttpcloudeventspydanticpydantic-settingsdefusedxmlazure-identitynumpyopenaiopenapi_corewebsocketsaiortcopentelemetry-apiopentelemetry-sdkprancepybars4jinja2nest-asyncioscipytyping-extensionsmcp |
| Maintenance | Actively maintained 8 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 2,765,519 / month, #2,897 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 5 - Production/StableFramework :: Pydantic :: 2Intended Audience :: DevelopersLicense :: OSI Approved :: MIT LicenseProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Typing :: Typed |
Evidence: semantic_kernel-1.44.1-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
An agent finds packages by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language. Give your agent the search over MCP.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also agent-framework-core · agent-framework-azure-ai · agent-framework-ollama · agent-framework · agent-framework-foundry-local · agent-framework-claude · marvin · agent-governance-toolkit-core · microsoft-agents-hosting-core · openai-agents