conductor

conductor

ConductorAgent — static-DAG planner (Sakana AI, arXiv 2512.04388).

Stage-1 inference-only repro. The paper's trained Qwen2.5-7B conductor is not released; we substitute a strong zero-shot cloud planner (default Opus) and run the same plan-then-execute machinery.

Pipeline per task:

1. Plan — the conductor reads the question + numbered worker pool and emits three lists (model_id, subtasks, access_list) in JSON, up to 5 steps.
2. Execute — for each step i: build the worker prompt from subtasks[i] + the concatenated prior (subtask, output) messages selected by access_list[i]; call worker model_id[i]; the final answer is the output of the last step.

On plan parse failure: retry once with a stricter "JSON only" prompt; on second failure, fall back to a single call to the strongest available worker (last in the pool by convention).

Workers come from cfg["workers"] or a sensible default pool (local Qwen if vLLM is up, plus Opus 4.7 and gpt-5-mini).

Hybrid harness result: conductor-swebenchverified-opusplan-30 = 0.367 acc / $0.22 per task — +10pp vs baseline-cloud at ~15× cheaper.

Ported from hybrid-local-cloud-compute/adapters/conductor_adapter.py.

Classes

ConductorAgent

ConductorAgent(*args: Any, **kwargs: Any)

Bases: LocalCloudAgent

Plan-then-execute static DAG over a worker pool. See module docstring.

Source code in src/openjarvis/agents/hybrid/conductor.py

def __init__(self, *args: Any, **kwargs: Any) -> None:
    super().__init__(*args, **kwargs)
    # Validate `method_cfg.worker_pool` early — surfaces config errors
    # at agent construction rather than on the first task. No-op when
    # the override is absent (default pool is built later, lazily,
    # because `_vllm_alive` needs a live network probe).
    if self._cfg.get("worker_pool") is not None:
        _resolve_worker_pool(
            self._cfg,
            self._local_model,
            self._local_endpoint,
            self._cloud_model,
        )

Functions