Train, deploy, and continuously improve your own models on an integrated compute, training, inference, and sandbox stack.
Visit primeintellect.ai → On Hacker News Oct 2024 – Aug 2026
AI & Machine Learning 4Software Engineering 2Tech Industry & Business 2
Prime Agent is our open-source, self-improving coding harness built around two abstractions: the Recursive Language Model (RLM) and the Continual Harness. With Opus 5, it achieves 95.5% on ARC-AGI-3, surpassing the reported human expert baseline.
We're excited to release INTELLECT-2, the first 32B parameter model trained via globally distributed reinforcement learning. Unlike traditional centralized training efforts, INTELLECT-2 trains a reasoning language model using fully asynchronous RL across a dynamic, heterogeneous swarm of compute contributors.
Compare 153 autonomous runs from 18 frontier models on the nanoGPT optimizer speedrun.
We're excited to launch INTELLECT-1, the first globally-distributed training run of a 10-billion-parameter model, inviting anyone to contribute compute and participate. This brings us one step closer towards open source AGI.