Companies / OpenPipe
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.5, GPT-OSS, Llama, and more!
Measured by us
98 observations · every few days
Where it is now
Over the 30 days to 31 Aug 2026 the repository added 120 stars. Across the 174 days we have recorded it averages 9.7 a day, so the last month is running at 0.4× that.
Finding companies at that stage is the product. Contributor pedigree, enterprise-demand signals and our ranking sit behind the login.
