{"data":{"slug":"openpipe-art","name":"ART","tagline":"Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO.","github_url":"https://github.com/OpenPipe/ART","owner":"OpenPipe","repo":"ART","owner_avatar_url":"https://avatars.githubusercontent.com/u/139012218?v=4","primary_language":"Python","stars":10603,"forks":974,"topics":["agent","agentic-ai","grpo","llms","lora","qwen","qwen3","reinforcement-learning","rl"],"archived":false,"github_pushed_at":"2026-08-19T01:31:53+00:00","maintenance_label":"Very active","stars_delta_30d":109,"url":"https://www.graphcanon.com/tools/openpipe-art","markdown_url":"https://www.graphcanon.com/tools/openpipe-art.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/openpipe-art","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=openpipe-art","description":"Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!","homepage_url":"https://art.openpipe.ai","license":"Apache-2.0","open_issues":127,"watchers":66,"ai_summary":"A framework for training multi-step AI agents in real-world tasks utilizing the GRPO method, compatible with various LLMs like Qwen3.6 and GPT-OSS.","readme_excerpt":"## Installation\n\nART agents can be trained from any client machine that runs python. To add to an existing project, run this command:\n\n```\npip install openpipe-art\n```\n\n---\n\n## ⚖️ License\n\nThis repository's source code is available under the [Apache-2.0 License](LICENSE).","github_created_at":"2025-03-10T18:25:47+00:00","created_at":"2026-07-07T17:39:22.449235+00:00","updated_at":"2026-08-19T12:01:02.347846+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"model-training","name":"Model Training","url":"https://www.graphcanon.com/categories/model-training","markdown_url":"https://www.graphcanon.com/categories/model-training.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/model-training"}],"tags":[{"slug":"agent","name":"agent"},{"slug":"agentic-ai","name":"agentic-ai"},{"slug":"grpo","name":"grpo"},{"slug":"llms","name":"llms"},{"slug":"lora","name":"lora"},{"slug":"qwen","name":"qwen"},{"slug":"reinforcement-learning","name":"reinforcement-learning"},{"slug":"rl","name":"rl"}],"trust":{"provenance":{"is_fork":false,"github_id":946178371,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-19T12:01:01.493Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":1,"days_since_push":0,"last_release_at":"2026-08-14T03:49:55Z","stars_delta_30d":109,"open_issues_delta_30d":6},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:14:23.549Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-19T12:01:01.976Z"},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-08-19T12:01:01.976Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-08-19T12:01:01.976Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-08-19T12:01:01.976Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["- When you need a framework that specifically supports the GRPO method for agent reinforcement learning.","- If your project involves integrating multi-step AI agents with real-world tasks, especially when using LLMs such as Qwen3.6 or GPT-OSS."],"when_not_to_use":["- Avoid ART if your training needs do not align with the GRPO method and you require a different approach for reinforcement learning.","- If your development revolves around non-supported Language Learning Models (LLMs), other than Qwen3.6, GPT-OSS, Llama etc., consider looking into alternative frameworks that offer broader model comp"],"source":"enrich:decision_facts","observed_at":"2026-07-11T15:47:34.173Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"ART is a specialized framework for training multi-step AI agents using GRPO, suitable for developers looking to train their agents with specific LLMs like Qwen3.6 and GPT-OSS."}]}}