--- title: "autogen vs raft" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/microsoft-autogen-vs-nvidia-raft" tools: ["microsoft-autogen", "nvidia-raft"] --- # autogen vs raft *GraphCanon updated Jul 11, 2026* ## Verdict Pick autogen when autogen is primarily Python; raft is Cuda; pick raft when raft is primarily Cuda; autogen is Python. [autogen](https://microsoft.github.io/autogen/) reports 60k GitHub stars, 9.0k forks, and 945 open issues, last pushed Apr 15, 2026. [raft](https://docs.rapids.ai/api/raft/stable/) has 1.0k stars, 240 forks, and 448 open issues, last pushed Jul 11, 2026. Figures are from public GitHub metadata via [autogen's repository](https://github.com/microsoft/autogen) and [raft's repository](https://github.com/NVIDIA/raft). | | [autogen](/tools/microsoft-autogen.md) | [raft](/tools/nvidia-raft.md) | | --- | --- | --- | | Tagline | A programming framework for agentic AI | RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing hig | | Stars | 59,658 | 1,026 | | Forks | 8,983 | 240 | | Open issues | 945 | 448 | | Language | Python | Cuda | | Adopt for | AutoGen is a Python-based framework for developing and managing agentic AI systems. It includes the AutoGen Studio for no-code GUI setup, integrating with various models. | - | | Persona | - | - | | Runtime | - | - | | License | CC-BY-4.0 | Apache-2.0 | | Categories | LLM Frameworks, AI Agents | LLM Frameworks, Vector Databases, Data & Retrieval | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [autogen](/tools/microsoft-autogen.md) | [raft](/tools/nvidia-raft.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Very active (96%) | | Days since push | 87d | 0d | | Open issues (now) | 945 | 448 | | Full report | [trust report](/tools/microsoft-autogen/trust.md) | [trust report](/tools/nvidia-raft/trust.md) | ## Decision facts: autogen - **Requirements:** Min 4 GB RAM; AutoGen requires Python 3.10 or later.; Ensure security when connecting to MCP servers due to the potential for local command execution and sensitive information exposure. - **Adopt for:** AutoGen is a Python-based framework for developing and managing agentic AI systems. It includes the AutoGen Studio for no-code GUI setup, integrating with various models. ## Choose when ### Choose autogen if… - autogen is primarily Python; raft is Cuda. - License: autogen is CC-BY-4.0, raft is Apache-2.0. - Requirements: Min 4 GB RAM; AutoGen requires Python 3.10 or later.; Ensure security when connecting to MCP servers due to the potential for local command execution and sensitive information exposure.. - Tags unique to autogen: llm-framework, autogen, agents, ai. - Also covers AI Agents. - You need a framework that supports integration with multiple AI models via OpenAI's chat completion client. ### Choose raft if… - raft is primarily Cuda; autogen is Python. - License: raft is Apache-2.0, autogen is CC-BY-4.0. - Tags unique to raft: clustering, anns, gpu, building-blocks. - Also covers Vector Databases, Data & Retrieval. ## When NOT to use autogen - If you require tools supporting multiple programming languages beyond Python, as AutoGen is strictly a Python-based framework. - When deploying in environments where connecting to external servers (like those used by MCP) could pose security risks or is prohibited. - You need solutions which do not involve additional installations for server components such as `playwright/mcp`, as AutoGen requires this setup for certain functionalities. ## When NOT to use raft - LLM Frameworks: Avoid a framework for a single prompt-and-retrieve call; the abstraction can cost more than it saves. - Vector Databases: Don't reach for a dedicated vector DB under ~100k vectors; pgvector on your existing Postgres is simpler to operate. - Data & Retrieval: Skip a heavy ingestion framework when your corpus is small and static; a script plus the embedding API is enough. ## Common questions ### What is the difference between autogen and raft? autogen: A programming framework for agentic AI. raft: RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing hig. See the comparison table for live GitHub stats and shared categories. ### When should I choose autogen over raft? Choose autogen over raft when autogen is primarily Python; raft is Cuda; License: autogen is CC-BY-4.0, raft is Apache-2.0; Requirements: Min 4 GB RAM; AutoGen requires Python 3.10 or later.; Ensure security when connecting to MCP servers due to the potential for local command execution and sensitive information exposure.; Tags unique to autogen: llm-framework, autogen, agents, ai; Also covers AI Agents; You need a framework that supports integration with multiple AI models via OpenAI's chat completion client. ### When should I choose raft over autogen? Choose raft over autogen when raft is primarily Cuda; autogen is Python; License: raft is Apache-2.0, autogen is CC-BY-4.0; Tags unique to raft: clustering, anns, gpu, building-blocks; Also covers Vector Databases, Data & Retrieval. ### When should I avoid autogen? If you require tools supporting multiple programming languages beyond Python, as AutoGen is strictly a Python-based framework. When deploying in environments where connecting to external servers (like those used by MCP) could pose security risks or is prohibited. You need solutions which do not involve additional installations for server components such as `playwright/mcp`, as AutoGen requires this setup for certain functionalities. ### When should I avoid raft? LLM Frameworks: Avoid a framework for a single prompt-and-retrieve call; the abstraction can cost more than it saves. Vector Databases: Don't reach for a dedicated vector DB under ~100k vectors; pgvector on your existing Postgres is simpler to operate. Data & Retrieval: Skip a heavy ingestion framework when your corpus is small and static; a script plus the embedding API is enough. ### Is autogen or raft more popular on GitHub? autogen has more GitHub stars (59,658 vs 1,026). Stars measure visibility, not whether either tool fits your constraints. ### Are autogen and raft open source? Yes - both are open-source projects on GitHub (autogen: CC-BY-4.0, raft: Apache-2.0). ### Where can I find alternatives to autogen or raft? GraphCanon lists graph-backed alternatives at [autogen alternatives](/tools/microsoft-autogen/alternatives) and [raft alternatives](/tools/nvidia-raft/alternatives) ([autogen markdown twin](/tools/microsoft-autogen/alternatives.md), [raft markdown twin](/tools/nvidia-raft/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/microsoft-autogen-vs-nvidia-raft.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, autogen or raft? autogen: Steady. raft: Very active. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for autogen and raft? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [autogen trust report](/tools/microsoft-autogen/trust); [raft trust report](/tools/nvidia-raft/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=microsoft-autogen`](/api/graphcanon/graph?tool=microsoft-autogen) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_