Recursive Language Models

TL;DR

Recursive Language Models (RLMs) enable large language models to process much longer prompts by recursively examining and decomposing input snippets, significantly improving performance on long-context tasks.

Contribution

The paper introduces RLMs, a novel inference paradigm allowing LLMs to handle inputs far beyond their context window through recursive processing, and presents the first RLM-based model, RLM-Qwen3-8B.

Findings

RLMs process inputs up to 100 times longer than standard models.

RLM-Qwen3-8B outperforms the base Qwen3-8B by 28.3% on average.

RLMs outperform vanilla LLMs and scaffolds on four long-context tasks.

Abstract

We study allowing large language models (LLMs) to process arbitrarily long prompts through the lens of inference-time scaling. We propose Recursive Language Models (RLMs), a general inference paradigm that treats long prompts as part of an external environment and allows the LLM to programmatically examine, decompose, and recursively call itself over snippets of the prompt. We find that RLMs can successfully process inputs up to two orders of magnitude beyond model context windows and, even for shorter prompts, dramatically outperform the quality of vanilla frontier LLMs and common long-context and coding scaffolds (e.g., on GPT-5 by a median across the evaluated benchmarks of $26\%$ against compaction, $130\%$ against CodeAct with sub-calls, and $13\%$ against Claude Code) across four diverse long-context tasks while having comparable cost. At a small scale, we post-train the first model around the RLM. Our model, RLM-Qwen3-8B, outperforms the underlying Qwen3-8B model by $28.3\%$ on average and even approaches the quality of vanilla GPT-5 on three long-context tasks. Code is available at https://github.com/alexzhang13/rlm.