agent-framework-lab
Experimental modules for Microsoft Agent Framework
Decision gist · record as of 2026-08-14
Yes, if you are actively experimenting with agent frameworks, benchmarking agent performance, or researching RL-based agent training. No, if you need stable, production-ready agent features—use the core framework instead. The package is well-maintained, has no known vulnerabilities, and carries low install friction, but its beta status and explicit warning about breaking changes mean you should only adopt it for non-critical experimentation.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.10 or later; agent-framework-core must be installed as a runtime dependency.
- Low install friction with a single runtime dependency.
- Active maintenance with recent releases and no known vulnerabilities.
License · maintenance · safety
permissive license (permissive) — MIT license (permissive) means you can use, modify, and distribute this package freely with minimal restrictions, though you must include the license notice.
last release 2026-07-10 (35 days) · last repo commit 2026-08-14 · 12,804 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 289,271 downloads/mo, #8,003 on PyPI
Alternatives
Verify before relying
pip install "agent-framework-lab[gaia]"
from agent_framework.lab.gaia import GAIA- Whether GAIA, TAU2, and Lightning modules are fully functional in the current beta release or still under active development.
- Whether breaking changes between releases are documented in a changelog.
- Performance characteristics and resource requirements for Lightning RL training and benchmark evaluation.
What it is and what it does
Agent Framework Lab is an experimental package from Microsoft that extends the core Agent Framework with research prototypes, benchmarking tools, and incubating features. It provides three main lab modules: GAIA for evaluating agents on general assistant tasks, TAU2 for customer support task evaluation, and Lightning for reinforcement learning agent training. The package is explicitly designed for experimentation and research rather than production use; its documentation warns that lab modules may experience breaking changes or deprecation without notice.
The package is installed via extras syntax (e.g., `pip install agent-framework-lab[gaia]`) and imported from the `agent_framework.lab` namespace. It depends only on agent-framework-core at runtime, keeping the dependency surface minimal. The beta status and active maintenance indicate ongoing development, with recent releases and community interest in the broader Agent Framework ecosystem.
Use it for
- Benchmark your agent implementations against GAIA or TAU2 evaluation suites to measure performance on general or customer support tasks.
- Prototype and experiment with new agent features in a lab environment before considering them for the core framework.
- Train agents using reinforcement learning with the Lightning module for custom task optimization.
- Evaluate different agent architectures and approaches in a research setting without production stability constraints.
- Contribute community-maintained lab modules by building on top of the core framework and Agent Framework Lab infrastructure.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are actively experimenting with agent frameworks, benchmarking agent performance, or researching RL-based agent training.
No, if you need stable, production-ready agent features—use the core framework instead. The package is well-maintained, has no known vulnerabilities, and carries low install friction, but its beta status and explicit warning about breaking changes mean you should only adopt it for non-critical experimentation.
Install
agent-framework-lab on PyPI
Before you install
Low install friction with a single runtime dependency. Active maintenance with recent releases and no known vulnerabilities. Beta status signals ongoing development; breaking changes are explicitly expected.
Requires Python 3.10 or later; agent-framework-core must be installed as a runtime dependency.
License in practice
MIT license (permissive) means you can use, modify, and distribute this package freely with minimal restrictions, though you must include the license notice.
Quickstart
pip install "agent-framework-lab[gaia]"
from agent_framework.lab.gaia import GAIAVerify before relying
- Whether GAIA, TAU2, and Lightning modules are fully functional in the current beta release or still under active development.
- Whether breaking changes between releases are documented in a changelog.
- Performance characteristics and resource requirements for Lightning RL training and benchmark evaluation.
Package facts
| License | permissive license permissive |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 1 packageagent-framework-core |
| Maintenance | Actively maintained 35 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 289,271 / month, #8,003 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: DevelopersLicense :: OSI Approved :: MIT LicenseProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Programming Language :: Python :: 3.14 |
Evidence: agent_framework_lab-1.0.0b260709-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “agent framework benchmarking”
- agent-framework-labProvides experimental lab modules (GAIA, TAU2, Lightning) built on…
- terminal-benchTerminal-Bench provides a benchmark suite and execution harness for…
- langwatch-scenarioLangwatch-scenario is an agent testing framework that simulates…
Give your agent the search over MCP, or paste the wish link into any chat.
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 agentlightning · agent-framework-foundry-local · agent-framework-declarative · agent-framework-ollama · agent-framework · agent-framework-core · agent-framework-claude · agent-framework-azure-ai · nemo-gym · agent-framework-devui