OpenPipe/ART
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!
Latest metrics (snapshot 2026-10-14)
| stars | 10801 |
|---|---|
| forks | 1007 |
| issues | 167 |
| pushed_at | 2026-10-12T20:55:34Z |
History
History chart: 1 daily snapshot so far (archive started 2026-10-14). The line fills in as the daily archive grows; full series via the API.
| Date | GitHub stars |
|---|---|
| 2026-10-14 | 10801 |
ADC Momentum: 52/100
| Growth velocity | 55 |
|---|---|
| Consistency | 44 |
| Recency | 91 |
| Agent demand | 0 |
| Basis | bootstrap |
ADC Momentum (0-100) = 45% growth velocity (stars/likes gained per day in our daily archive, relative and absolute, percentile-ranked within the category) + 20% consistency (share of recent days with growth) + 20% recency (latest push/update/release, decaying) + 15% agent demand (how often AI agents looked it up via our API/MCP this week). MCP servers use release cadence; LLMs use price cuts, context and recency. Recomputed daily; scores marked "bootstrap" use lifetime rates until the archive has enough days. Method
Embed this badge
Maintainers: add the live badge to your README (updates daily).
[](https://agentdataco.com/repo/OpenPipe/ART)<a href="https://agentdataco.com/repo/OpenPipe/ART"><img src="https://agentdataco.com/badge/repo/OpenPipe/ART.svg" alt="ADC Momentum"></a>Details
| 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! |
|---|---|
| url | https://github.com/OpenPipe/ART |
| language | Python |
| license | Apache-2.0 |
| topics | agent agentic-ai grpo llms lora qwen qwen3 reinforcement-learning rl |
| created_at | 2025-03-10T18:25:47Z |
| homepage | https://art.openpipe.ai |
Get this data via API or MCP
Full daily history, trends and change events for OpenPipe/ART and thousands of other AI-ecosystem entities.
curl -H "Authorization: Bearer $ADC_KEY" \
"https://agentdataco.com/api/v1/github-ai-repos/history?entity=OpenPipe%2FART"
curl -H "Authorization: Bearer $ADC_KEY" \
"https://agentdataco.com/api/v1/github-ai-repos?repo=OpenPipe%2FART"Or connect your agent to our remote MCP server: https://agentdataco.com/mcp.
Maintainer? Claim this listing
Owners and vendors can submit a description, pricing and corrections. We email you a verification link; verified submissions are reviewed daily (a work email on the project's domain is fastest).