October 8, 2025 — Leads & Copy — CoreWeave, Inc. (Nasdaq: CRWV) has launched Serverless RL, a new tool designed to facilitate the training of AI agents using reinforcement learning (RL). According to the company, Serverless RL scales to dozens of GPUs and requires only a Weights & Biases account and API key to operate.
The release follows CoreWeave’s recent acquisition of OpenPipe. The company says its Serverless RL helps remove infrastructure constraints, allowing enterprises to improve AI agents and customer experience.
“Being fast to market is critical, and equally important is the elegance and ease of use we are now giving AI pioneers across labs, enterprises, and startups to fine-tune large language models and build AI agents with confidence,” said Peter Salanki, Co-founder and Chief Technology Officer of CoreWeave.
Benchmarks reportedly indicate nearly 1.4x faster training times and 40 percent lower costs compared to local H100 GPU environments, without affecting model quality. Customers such as SquadStack.ai and QA Wolf are already expressing interest in Serverless RL, according to CoreWeave.
QA Wolf CEO Jon Perl said they are eager for instant GPU access without infrastructure management, to improve the quality of their agents.
CoreWeave has added capabilities to its platform through acquisitions of OpenPipe, Weights & Biases, and Monolith.
To learn more about Serverless RL, visit www.wandb.ai/site/serverless-rl.
Peter Salanki, Co-founder and Chief Technology Officer of CoreWeave.
Source: CoreWeave
