Paper page - Prime Agent: A Self-Improving RLM Harness

Source: https://huggingface.co/papers/2608.23552 Authors:

,

,

,

,

,

,

,

,

,

Abstract

Prime Agent is an open-source harness that uses recursive subagents, persistent computation, and agent-to-agent coordination to extend language models’ long-horizon capabilities across coding and reasoning tasks.

Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context. Prime Agent is an open-source harness for long-horizon evaluation and coding-agent workflows. A persistent IPython REPL follows theRecursive Language Modelabstraction for programmatic context processing andtest-time compute, while Continual Harness preserves histories, memories, skills, prompts, and subagent specifications across trajectories. Recursivesubagentscoordinate through directagent-to-agent communication, and the Agents View lets humans inspect and managedaemon-backed sessions. Prime Agent standardizes execution, recovery, verification, and resource accounting while leaving strategy construction to the model. This low-friction, expressive membrane prevents harness failures from becoming model failures and pushes measurement toward the model’s true maximal underlying capability. Prime Agent raisesARC-AGI-3RHAE Best@1 from 30% to 95.5% and matches or exceeds native and popular harnesses acrosslong-context coding,GPU-kernel generation, emulator construction, and autonomousnanoGPTspeedruns. On Factorio, we find refinement allows for continuous technology progression and dedicatedsubagentsenable parallelized work. Code is available at https://github.com/PrimeIntellect-ai/prime-agent.

View arXiv pageView PDFProject pageGitHub18.2kAdd to collection

Get this paper in your agent:

hf papers read 2608\.23552

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.23552 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.23552 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.23552 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.