|

Prime Intellect Releases Prime Agent: An Open-Source RLM Harness Where Sub-Agents Are Function Calls Inside Persistent IPython Kernel

ByRicardo

Prime Intellect has open-sourced Prime Agent, a self-improving coding harness designed round two abstractions, the Recursive Language Model (RLM) and Continual Harness. Fixed software schemas and context compaction power a mannequin to work round its personal scaffolding. Prime Agent replaces each with a persistent Python REPL and a rewritable harness. With Opus 5, it stories 95.5% on ARC-AGI-3, above the reported human knowledgeable baseline of 95.4%. It is MIT-licensed.

Is it deployable

Yes, immediately. Prime Agent installs on Linux or macOS with one command. It runs on subscription logins (Codex, Claude Pro/Max, GitHub Copilot), API keys (Anthropic, OpenAI, Google, Groq, Fireworks, Prime Inference, and others), Azure OpenAI, Amazon Bedrock, and self-hosted vLLM, Ollama, or LM Studio endpoints. Self-hosting an open-weights mannequin resembling GLM-5.2 retains code inside your personal community.

  • Company degree: Best match is mid-size to giant engineering orgs and AI labs that already run remoted CI containers. Prime Intellect states plainly that employee and kernel processes are not a safety sandbox. Deployment subsequently wants disposable clones or restricted environments. Solo builders can set up it, however the payoff seems on multi-hour duties.
  • Industries: Developer tooling, semiconductor and HPC groups writing GPU kernels, simulation and gaming, quantitative analysis, and AI analysis labs.
  • Applications: Overnight refactors behind a check gate, spec-driven builds from scratch, kernel optimization, long-horizon agent analysis, and autoresearch.

What Prime Intellect shipped

Prime Agent is constructed on two abstractions. The Recursive Language Model (RLM) treats context as a variable and sub-agent delegation as operate calls inside a REPL. The Continual Harness treats prompts, sub-agents, expertise, and reminiscence as state the agent can create, learn, replace, and delete from its personal trajectory. Both papers have Prime Agent authors on them. The TUI is constructed on pi.