Train a model using reinforcement learning on Prime Intellect’s infrastructure — no GPUs to manage.
Shared Hosted Training for LoRA runs will stop accepting new runs on October 5, 2026. Hosted Training is transitioning to dedicated runs. Existing LoRA adapters will remain downloadable and deployable until further notice.
The steps below cover the legacy shared LoRA service. Full fine-tuning (FFT) is currently in closed beta.
That’s it. Prime Intellect handles the GPU cluster, rollouts, and optimization. You’ll get a dashboard link in the output.You can also launch your coding agent (Claude, Cursor, OpenCode, etc.) and use it to manage your workspace and research workflows.
