> ## Documentation Index
> Fetch the complete documentation index at: https://docs.qredence.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction to fleet-rlm

> fleet-rlm is a Daytona-backed web workspace for recursive language-model tasks on DSPy, with a ReAct chat agent that delegates long-context work to dspy.RLM.

`fleet-rlm` is a web workspace for running **recursive language-model tasks** on top of [DSPy](https://dspy.ai) and [Daytona](https://daytona.io) sandboxes. You chat with a ReAct agent in the browser; when a task is larger than a single context window, the agent delegates pieces to isolated sub-sandboxes, each running a bounded `dspy.RLM` per [arXiv 2512.24601v2](https://arxiv.org/abs/2512.24601).

## Who it's for

DSPy users who want a UI-driven workspace for long-context tasks, recursive decomposition, and sandboxed code execution — without hand-rolling the transport, persistence, and sandbox plumbing.

## What it removes

Writing your own WebSocket transport, session persistence, Daytona sandbox lifecycle, execution-trace UI, and recursive-delegation policy around a DSPy program. `fleet-rlm` ships all of that behind a single `uv run fleet web`.

## Two layers, both `dspy.*`

<CardGroup cols={2}>
  <Card title="Chat surface" icon="comments">
    `dspy.ReAct` for interactive turn-taking. Implemented as `FleetAgent` in `src/fleet_rlm/runtime/agent/agent.py`.
  </Card>

  <Card title="Recursive engine" icon="diagram-project">
    `dspy.RLM` running inside a child Daytona sandbox. Built in `src/fleet_rlm/runtime/models/builders.py` via `build_recursive_subquery_rlm()`.
  </Card>
</CardGroup>

The chat agent is the entry point. The recursive engine runs when a task exceeds what a single ReAct context can handle. Both share a single LLM-call budget across the recursive tree.

## Empirical capability

Fleet-RLM's RLM capabilities were benchmarked against the published [RLM paper](https://arxiv.org/abs/2512.24601) and Prime Intellect's official `oolong-rlm` environment.

| Benchmark                              | Paper RLM (GPT-5) | Fleet-RLM + Gemini 3.1 Pro |
| -------------------------------------- | ----------------- | -------------------------- |
| S-NIAH (50 tasks, 50K–200K chars)      | (solved)          | **100.0%**                 |
| OOLONG-Official (`trec_coarse` @ 128K) | **56.5%**         | **91.67%** (+35.2 pp)      |
| OOLONG synthetic (30 tasks)            | 56.5% (reference) | 74.0%                      |

## Where to go next

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/fleet-rlm/quickstart">
    Install fleet-rlm and launch the Web UI in 30 seconds.
  </Card>

  <Card title="Architecture" icon="sitemap" href="/fleet-rlm/concepts/architecture">
    Thin transport, runtime core, and Daytona substrate.
  </Card>

  <Card title="Agent model" icon="robot" href="/fleet-rlm/concepts/agent-model">
    `FleetAgent` (`dspy.ReAct`), `AgentRuntime`, signatures, and core memory.
  </Card>

  <Card title="Recursive RLM" icon="diagram-project" href="/fleet-rlm/concepts/recursive-rlm">
    Algorithm 1, delegation, REPL-variable mode, and the shared call budget.
  </Card>

  <Card title="Daytona runtime" icon="cubes" href="/fleet-rlm/concepts/daytona-runtime">
    Sandbox lifecycle, volumes, session continuity, and the host-callback bridge.
  </Card>

  <Card title="Sessions & persistence" icon="database" href="/fleet-rlm/concepts/sessions-persistence">
    Manifests, stateful restore, and the Neon-backed tenant store.
  </Card>

  <Card title="Observability" icon="chart-line" href="/fleet-rlm/concepts/observability">
    MLflow tracing, WebSocket events, diagnostics, and trace feedback.
  </Card>

  <Card title="Python API" icon="code" href="/fleet-rlm/reference/python-api">
    `DaytonaInterpreter`, `FleetAgent`, runners, and DSPy signatures.
  </Card>
</CardGroup>

## Source of truth

When the docs disagree with the code, trust the code:

* Backend routes and WebSocket behavior live in `src/fleet_rlm/api/`.
* Runtime and Daytona execution live in `src/fleet_rlm/runtime/` and `src/fleet_rlm/integrations/daytona/`.
* The canonical HTTP schema is [`openapi.yaml`](https://github.com/Qredence/fleet-rlm/blob/main/openapi.yaml).