--- schema: formation.pattern/v0.1 kind: pattern visibility: public aliases: - 'dynamic subagents' - 'spawn-on-demand workers' - 'runtime agent creation' - 'subagent spawning' canonical_url: https://topologyindex.com/patterns/dynamic_spawning.md description: 'New workers are created during the run in response to what the run finds.' executable_here: false executor_requirements: - role_creation_at_runtime - concurrent_role_invocations family: adaptive findings_page: /patterns/dynamic_spawning/findings.md observed_in: [] path: /patterns/dynamic_spawning.md pattern: dynamic_spawning pattern_index: /patterns/index.md product_api_version: v1 published_findings: - benchmarks: [] compared_against: 'A single agent using the stronger model alone' direction: helped source_id: web:anthropic.com/engineering/multi-agent-research-system task_domain: 'Open-ended web research' task_domains: - research url: https://www.anthropic.com/engineering/multi-agent-research-system - benchmarks: [] compared_against: 'A single agent' direction: mixed source_id: arxiv:2308.10848 task_domain: 'Text understanding, reasoning, coding, tool use and embodied tasks' task_domains: - coding - reasoning - operations url: https://arxiv.org/abs/2308.10848 - benchmarks: [] compared_against: 'Existing multi-agent methods with predefined agents' direction: helped source_id: arxiv:2309.17288 task_domain: 'Open-ended question answering and creative writing' task_domains: - reasoning url: https://arxiv.org/abs/2309.17288 - benchmarks: [] compared_against: 'A single-threaded linear agent sharing full context' direction: hurt source_id: web:cognition.com/blog/dont-build-multi-agents task_domain: 'Software engineering agents' task_domains: - coding url: https://cognition.com/blog/dont-build-multi-agents - benchmarks: - Bamboogle - Musique - HotpotQA compared_against: 'The same model without search, a ReAct-style search agent, and the ChatGPT-Web and Perplexity Pro products' direction: helped source_id: arxiv:2407.20183 task_domain: 'Web information seeking and multi-hop question answering' task_domains: - research url: https://arxiv.org/abs/2407.20183 qualifying_evidence: [] references: - https://www.anthropic.com/engineering/multi-agent-research-system schema_version: v0.1 title: 'When to use the dynamic_spawning multi-agent pattern (dynamic subagents, spawn-on-demand workers)?' unmet_executor_requirements: - role_creation_at_runtime - concurrent_role_invocations --- # When to use the dynamic_spawning multi-agent pattern (dynamic subagents, spawn-on-demand workers)? **Short answer:** when the task has subtasks unknown until it starts, start with [supervisor](/patterns/supervisor.md) and consider it next. A hypothesis from the [decision guide](/patterns/index.md), not a ranking; [published findings](/patterns/dynamic_spawning/findings.md) keep the unfavourable ones. The `dynamic_spawning` multi-agent pattern (family `adaptive`): new workers are created during the run in response to what the run finds. Also called: dynamic subagents, spawn-on-demand workers, runtime agent creation, subagent spawning. ## The arrangement - Structure: Roles that do not exist until something in the run asks for them. - Control: A spawning decision taken inside the run. - Shared state: Whatever the spawning role hands to what it spawned. ## When it is hypothesised to fit Hypothesised to fit when the shape of the work is unknown until it starts, so a fixed role set is either wasteful or too small. A hypothesis to test on a workload, not a finding: this service has not tested it. ## Failure modes to watch for - A spawn decision is a spending decision; if a model makes it, something else must hold the budget. - What ran is no longer what was registered, so an evaluation cannot say which arrangement produced the result. - Runaway spawning is the failure this arrangement is bounded against. These are things to watch for, not outcomes anyone measured here. ## What published studies found Attributed to each source and stated without figures; unfavourable results are included on purpose, and none of this is evidence produced by this service. What each was compared with is in `published_findings`; each finding in words, with its caveat, is at [/patterns/dynamic_spawning/findings.md](/patterns/dynamic_spawning/findings.md). Reviewed 2026-09-23. Each label says how this pattern fared against what it was compared with, as the source reports it; the labels are defined at [/docs/schemas/pattern/v0.1.md](/docs/schemas/pattern/v0.1.md). - **Helped**: [Anthropic Engineering: How we built our multi-agent research system](https://www.anthropic.com/engineering/multi-agent-research-system) - **Mixed**: [AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors](https://arxiv.org/abs/2308.10848) - **Helped**: [AutoAgents: A Framework for Automatic Agent Generation](https://arxiv.org/abs/2309.17288) - **Hurt**: [Cognition blog: Don't Build Multi-Agents](https://cognition.com/blog/dont-build-multi-agents) - **Helped**: [MindSearch: Mimicking Human Minds Elicits Deep AI Searcher](https://arxiv.org/abs/2407.20183) ## Sources that describe it Where the arrangement is described. A source is a reason to test it, never a result for it. - [How we built our multi-agent research system (Anthropic)](https://www.anthropic.com/engineering/multi-agent-research-system) ## Can this deployment execute it No. The runner does not provide everything this arrangement needs. What is missing: - `role_creation_at_runtime` (roles that do not exist until the work asks for them): Roles come from the registered formation; the supervisor can only move between declared roles. - `concurrent_role_invocations` (more than one agent acting at the same time): The supervisor runs one invocation at a time and refuses a declared active-agent count above one, rather than serializing a parallel policy and calling it the same thing. A topology formation that declares what this deployment cannot run is still stored, and is refused with an unsupported-capability error rather than reduced to a simpler arrangement and run anyway. ## Evidence for this pattern None. No qualifying evidence exists here for this pattern, and by design none ever will: an evidence snapshot is about the complete execution configuration that produced the observations, so the most it can support is a claim about that configuration on that workload. ## Observed in No research record on this service reports observing this arrangement. ## How evidence works here A pattern names how work is organized, never how well that works. Evidence in this service attaches to a complete execution configuration — a topology formation together with an execution lock pinning models, tools and runtime — and never to a pattern on its own, so no page here carries a success rate, a cost or a run count for an arrangement.