RLMs: an approach to very long prompts
Published on October 15, 2025, the publication describes language models that interact recursively with long prompts through a REPL and reports evaluations on the OOLONG and BrowseComp-Plus benchmarks.
Source: Recursive Language Models processam prompts longos por meio de um REPL (x.com). Text prepared with AI from this source.
What happened and what to do
On October 15, 2025, the publication described Recursive Language Models (RLMs), an inference strategy in which language models recursively decompose and explore very long prompts through a REPL. It reports results on the OOLONG and BrowseComp-Plus benchmarks and presents the approach as a possible alternative to explicit retrieval for long-context tasks.
A company can evaluate this strategy in workflows that depend on extensive documents, comparing it with explicit retrieval on tasks representative of its business. This may involve preparing evaluation sets, instrumenting calls and costs, measuring quality and latency, and setting security controls before integrating the processing into an internal application.
How the consultancy can help
Wendelmaques can diagnose use cases involving extensive contexts, define a comparative evaluation, and scope implementation of a workflow with monitoring, controls, and operation on the company's infrastructure.
Next step
Send a short description of your use case to receive a proposal scoped for diagnosis and implementation.
Consulting for your project
Infrastructure review, deployment and ongoing operations, with scope and pricing defined in the proposal.
Quoted per project
Request a proposal