Multi-agent: a census of 143 listings

Two named workers, Alice on outbound and Julian on inbound, run on one platform and one API, but the vendor never describes several agents coordinating on a single task.

Source retrieved 2026-09-23

Corpable states it builds bespoke multi-agent workflows and ships twelve separate agent packs, but no page describes agents coordinating with each other on a single task or any orchestration mechanism.

Source retrieved 2026-09-27

Coordinating two agents on one task is the entire product: both run as persistent peers, either side can push mid-turn, and they negotiate their own task splits.

Source retrieved 2026-10-03

Multi-agent builds are an advertised deliverable, with a supervising orchestrator coordinating agents and escalating to people, listed among the firm's named specialties.

Source retrieved 2026-10-03

Four coordination topologies ship as runtime factories alongside directed delegation, so a planner can route work to named specialist runtimes on one task.

Source retrieved 2026-09-09

Teams coordinate several agents in one of four modes; workflows sequence agents and teams across six step types.

Source retrieved 2026-09-02

One loadout is reusable by several agents, clients and devices at once, but the vendor documents no orchestration of multiple agents cooperating on a single task.

Source retrieved 2026-10-03

Architect mode splits one task across two model roles, an architect that proposes and an editor that writes the edits, which is role separation rather than several agents coordinating freely.

Source retrieved 2026-09-30

Native orchestration between domain-specific agents is shipped and documented; coordination with third-party agents over A2A is a separate, unreleased capability held in Aisera Unify.

Source retrieved 2026-09-03

The vendor names several distinct agents in one pipeline - scanning for signals, researching context, then writing sequences - but never describes them coordinating with or delegating to each other, and there is no orchestration or custom-agent builder documented.

Source retrieved 2026-09-10

The main agent spawns concurrent subagents with their own tool grants, monitored in an /agents panel; collaborative agent teams via /teamwork-preview are gated to paid plans.

Source retrieved 2026-09-10

Several agents can share one workspace's lists and credit balance with a key each, so spend is attributable per agent, but Argorant supplies no orchestration or hand-off between them.

Source retrieved 2026-09-26

The vendor describes swarms of research agents used for enrichment and qualification, which is one orchestrated research step rather than several named agents you compose or assign to a task.

Source retrieved 2026-09-10

The AI Agents framework is explicitly multi-agent: several specialist agents collaborate on one call and callers are routed between them, with documented multi-agent topologies.

Source retrieved 2026-09-24

A manager Expert launches worker Experts or subagents that run in parallel with their own context and model.

Source retrieved 2026-09-02

Coordination of several agents is the core design: a lead orchestrator decomposes tasks and assigns subtasks to web, file, coder and terminal agents.

Source retrieved 2026-09-19

Multi-agent systems are named as an integration target, and the build stack cites AutoGen Studio and CrewAI.

Source retrieved 2026-09-02

Warm transfer runs a second proxy-agent leg alongside the primary agent, but that is call handoff, not general orchestration.

Source retrieved 2026-09-11

You can create unlimited agents on one account, but the vendor's own comparison against CrewAI concedes it does little role-based orchestration, so several agents cooperating on a single task is not a published capability.

Source retrieved 2026-09-13

Thousands of sessions run in parallel and Stagehand plugs into multi-agent frameworks; Browserbase itself does not orchestrate agents.

Source retrieved 2026-09-02

Subagents are first-class: defined in code or as files, run concurrently, each with its own context and tool subset.

Source retrieved 2026-09-02

Four coordination styles: subagents, background agent view, agent teams with inter-agent messaging (experimental, off by default), and dynamic workflows.

Source retrieved 2026-09-02

Two mechanisms coordinate several agents: read-only subagents that research in parallel, and SDK agent teams where a coordinator delegates to specialists and merges results.

Source retrieved 2026-09-12

The vendor describes several specialised agents running in parallel inside one review, rather than a single prompt, plus separate chat and finishing-touch agents.

Source retrieved 2026-09-15

Spawns parallel subagents and consolidates their results; /agent switches between threads.

Source retrieved 2026-09-02

Seven named agents divide the operation — Concierge, Quality, Agent Assist, Applied Insights, Optimization, Integration and Knowledge — and feed each other in a documented improvement loop.

Source retrieved 2026-09-12

Agents are built as a decentralised network of specialised subAgents, each with its own prompts, knowledge, tools and guardrails, collaborating on one conversation.

Source retrieved 2026-09-13

Multi-agent coordination is the core abstraction: crews of specialised agents delegate between themselves, with MCP for tooling and A2A for cross-agent discovery.

Source retrieved 2026-09-13

Subagents run in parallel in their own context windows in editor, CLI and cloud; /multitask and Build in Parallel fan a plan out across them.

Source retrieved 2026-09-02

Several specialised models act on one conversation — a supervisor checking responses and a bad-actor screen — but Decagon never describes customers composing or orchestrating multiple agents themselves.

Source retrieved 2026-09-14

Handoff between specialised agents is a documented pattern with a reference implementation you run yourself: your orchestrator opens a fresh Voice Agent session per agent and summarises context between them. The API itself has no built-in agent-to-agent coordination.

Source retrieved 2026-09-02

A coordinator Devin delegates to child sessions running in parallel, each on its own isolated VM.

Source retrieved 2026-09-02

Local, cloud and third-party ACP agents run side by side in the Kanban Agent Command Center, Spaces share task context between them, and the local agent spawns its own subagents.

Source retrieved 2026-09-03

Agents can be placed as nodes inside a workflow so several specialised agents hand work to each other.

Source retrieved 2026-09-02

Multi-agent orchestration with scheduled and event-driven triggers is listed for both Business and Enterprise, and sub-agent credit costs accumulate across an agent chain.

Source retrieved 2026-10-01

An orchestrator agent routes a live conversation to specialist agents on natural-language conditions, and transfers can be nested; the transcript carries across.

Source retrieved 2026-09-02

Task-launched subagents run in fresh contexts, concurrently or in background; Missions orchestrate workers and validators.

Source retrieved 2026-09-02

Two agents exist and one acts on the other: Operator tunes, tests and ships changes to Fin, but Operator sits in the paid Pro add-on and its changes are proposals you approve rather than joint work on one customer conversation.

Source retrieved 2026-09-18

One agent run spans an orchestrator plus its own subagents, each emitting trace events under a separate agent id. But that topology is internal and undocumented as a feature: there is no parameter to compose, name or budget the subagents.

Source retrieved 2026-09-03

Chat 2.0 runs specialised research and analysis processes in parallel and reconciles their findings before answering, so coordination happens inside one conversation rather than as user-assembled agents.

Source retrieved 2026-09-14

GenTeam puts several named agents in shared channels with humans, agents claim tasks and hand results to each other, and Custom Agents can be combined inside one Super Agent conversation.

Source retrieved 2026-09-15

Copilot CLI delegates to specialised sub-agents on every plan; delegating a task to Claude or Codex is a Pro+ and above preview, and the CLI's ACP server is documented for coordinating with other agents.

Source retrieved 2026-09-02

Orchestration activates specialist agents in sequence on one multi-step task and shares context between them.

Source retrieved 2026-09-02

Multi-agent composition is the headline design: hierarchies of specialised agents that delegate sub-tasks, plus sequential, parallel and loop workflow agents for deterministic orchestration.

Source retrieved 2026-09-15

goose spawns subagents that run sequentially or in parallel with their own process isolation, driven either by direct prompts or by reusable recipe files, and can also wrap external agents such as Codex.

Source retrieved 2026-09-16

A LangGraph flow coordinates a chief editor, researcher, reviewer, revisor, writer and publisher, and an AG2 integration is documented too.

Source retrieved 2026-09-16

Multiple agents can be orchestrated inside deterministic workflows and an agent can delegate to subagents, so coordination is a first-class feature.

Source retrieved 2026-09-30

One deployment carries several specialised agents (support, IT help desk, knowledge, ticket automation, QA), but the vendor never documents several agents coordinating on a single conversation.

Source retrieved 2026-09-03

Work is delegated to nested sub-threads that run in isolation and return only the requested result to the parent, and read-only strategies drive Explore/Research sub-agents; coordination is parent-to-child, not peer agents on a shared bus.

Source retrieved 2026-09-17

One background sub-agent exists and runs after conversations to learn your schema, but the vendor documents no way for a user to orchestrate several agents on a single task.

Source retrieved 2026-09-17

Custom subagents run as independent contexts that the main agent selects and delegates to, with their own model, tool restrictions and parallel execution across subtasks.

Source retrieved 2026-09-18

Orchestrator mode coordinates other modes; the CLI and JetBrains plugin run parallel agents in isolated worktrees.

Source retrieved 2026-09-02

LangGraph is built for coordinating several agents in one graph, including hierarchical supervisors, and dcode delegates work to task-specific subagents in parallel.

Source retrieved 2026-09-18

One agent can call another agent as a tool, so several agents coordinate inside a single flow; the project advertises multi-agent orchestration as a headline feature.

Source retrieved 2026-09-25

Single, multi-agent and hierarchical control flows are all expressed in the same framework; the LangChain docs enumerate subagent, handoff, router and custom-workflow patterns.

Source retrieved 2026-09-02

A fixed vendor pipeline rather than a crew you compose: separate agents handle discovery, enrichment and scoring, a drafting agent writes, and a second model proofreads every message before review.

Source retrieved 2026-09-18

Seven built-in subagent types plus deploying any existing stateful agent as a subagent, in parallel or background.

Source retrieved 2026-09-02

Lindy Build pairs the builder with separate QA agents working the same app; the Teammate product gives each person a private Lindy plus one shared team Lindy, and the vendor describes no orchestration between them.

Source retrieved 2026-09-02

Multiple agents in one session via handoffs, plus a supervisor pattern where one agent keeps control and delegates to scoped tasks.

Source retrieved 2026-09-02

Workflows orchestrates multi-step systems and can combine several agents in one run.

Source retrieved 2026-09-02

Specialised agents run under a supervisor with delegation, shared state and conflict resolution, and the workflow product names LangGraph, Autogen and CrewAI as the orchestration underneath.

Source retrieved 2026-09-09

Wide Research decomposes one request into independent sub-tasks and runs many agents in parallel, each with its own context, then a main agent synthesises the results.

Source retrieved 2026-09-19

Supervisor/subagent delegation is a first-class primitive, with delegation hooks, message filtering and per-subagent memory isolation.

Source retrieved 2026-09-02

Maven describes several outcome-specific agents on one platform and per-situation Charters inside a single reasoning engine, but it explicitly avoids channel-specific or peer agents coordinating with each other, so this is one governed agent with many roles rather than a multi-agent orchestration graph.

Source retrieved 2026-09-25

Five built-in orchestration patterns coordinate several agents on one task, and a parent agent can additionally delegate concurrent work to named child agents in their own sessions.

Source retrieved 2026-09-19

A workflow can spawn one or more child workflows, and packaged agents can be called as blocks inside another agent, giving multi-agent orchestration inside one run.

Source retrieved 2026-09-20

You can run several AI Agents with different roles and route conversations to each, but the docs describe one agent per conversation assignment rather than agents coordinating on one task.

Source retrieved 2026-10-03

The reasoning engine selects and chains multiple plugins in one conversation, and ambient agents can hand work to conversational agents on the same platform.

Source retrieved 2026-09-20

A lead session spawns child agents that can themselves spawn grandchildren, with eight concurrent agents per tree by default and a configurable ceiling of 64.

Source retrieved 2026-09-20

A supervising agent delegates to specialised agents as tools, nestable across layers; published agents can also be attached as sub-agents.

Source retrieved 2026-09-02

Workflows chain agent actions and can call sub-workflows, so several agents run in sequence, but the docs describe no agent-to-agent handoff or shared-state orchestration.

Source retrieved 2026-09-09

Coordination is a core primitive: agents can hand off the conversation to a specialist, or an orchestrator can call other agents as tools within one run.

Source retrieved 2026-09-20

A RealtimeAgent accepts handoffs like a text agent, so spoken workflows can branch across specialist agents in one session.

Source retrieved 2026-09-02

One Gateway can run several isolated agents with their own workspaces and sessions, and a bundled team preset wires a coordinator that delegates to researcher, writer and reviewer roles.

Source retrieved 2026-09-21

Primary agents delegate to subagents in child sessions, and the vendor recommends it explicitly for running work in parallel.

Source retrieved 2026-09-21

A parent agent can delegate to sub-agents, but the documented TaskToolSet runs them sequentially and blocks, so it is delegation rather than parallel coordination.

Source retrieved 2026-09-21

Computer fans a prompt out to subagents; the API's wide-research preset runs the same wide-and-deep pattern in the background.

Source retrieved 2026-09-02

Agents can be chained: one agent is connected as another's Action so a triage agent routes to specialists, which is also the vendor's recommended way past its four-action guidance.

Source retrieved 2026-09-09

The SubAgents capability gives a parent agent a delegate_task tool over named child agents, each running in its own message history with its own budgets, which is coordination between several agents on one task.

Source retrieved 2026-09-21

Running more than one agent persona at once is a plan-gated capability: multiple agent profiles appear only in the top Ultimate tier's feature list, not in Premier or Enterprise.

Source retrieved 2026-09-22

Workforce is a visual canvas for wiring several specialist agents together, and Workforces are unlimited on every published plan tier.

Source retrieved 2026-09-22

Several tasks run in parallel with automatic dependency detection and AI-assisted conflict resolution, though the ceiling varies sharply by plan.

Source retrieved 2026-09-22

One call can be handed between several specialised AI agents mid-conversation, with the destination agent inheriting the full transcript, so work is split across agents rather than one monolith.

Source retrieved 2026-09-23

An orchestrator builds pipelines of specialised worker agents (research, email composition, verification) per query; Rox brands the layer Agent Swarm.

Source retrieved 2026-09-03

Several concurrent agent workflows and multi-channel agent teams are offered only in the top Multi-System Custom Scope build, not in the entry channel-neutral or WhatsApp packages.

Source retrieved 2026-09-28

A primary agent delegates to secondary specialist agents routed by the Atlas Reasoning Engine, and can delegate to third-party agents over Agent2Agent.

Source retrieved 2026-09-03

Several MCP clients can share one SalesTouch connection and workspace state; no orchestration of multiple agents on a single task is described.

Source retrieved 2026-09-03

The vendor describes multiple coordinating agents running hundreds of tasks per article, and markets the product as a set of SEO AI agents rather than one model call.

Source retrieved 2026-09-25

Several named role agents (Strategist, Content Writer, Technical SEO, On-Page Optimizer, Link Builder) run one approved campaign in sequence, managed from a single board.

Source retrieved 2026-10-03

Two vendor agents hand work to each other — Explorer's findings go straight to Ghostwriter — but customer-built agents coordinating on one task is not documented publicly.

Source retrieved 2026-09-26

Multiple agent blocks run in one workflow, and the Parallel container executes branches concurrently.

Source retrieved 2026-09-03

One flow owns a call; coordination between agents happens by transferring the call with context passing rather than by any orchestration framework.

Source retrieved 2026-09-03

One audit job can be scoped to several specialist reviewers at once, which is parallel specialists on a shared input rather than agents negotiating or delegating between themselves.

Source retrieved 2026-10-03

A supervising model runs alongside the conversational agent on the same live call, and paid tiers allow three to ten separate agents on one account.

Source retrieved 2026-09-26

The assistant can fan research out to parallel sub-agents, but those helpers are read-only and stateless, so several agents do not jointly write or coordinate a task.

Source retrieved 2026-09-15

Subflow Tools let one orchestrating agent call other agents as named specialists, in parallel or in sequence, each with its own workflow and output node.

Source retrieved 2026-09-09

Subflows act as specialised agents that a parent flow triggers for individual tasks.

Source retrieved 2026-09-03

Several specialised agents run side by side in one Agent-OS workspace with separate tools and memory, but no page describes agents delegating to each other or coordinating on a single task.

Source retrieved 2026-09-12

A handoff tool passes one live conversation between specialist assistants sharing context, and remote A2A agents plus multi-participant calls extend that beyond a single Telnyx assistant.

Source retrieved 2026-09-09

Two separate mechanisms: an agent can add another published agent as a tool and chain capabilities, and Maestro coordinates several agents, robots and people across one BPMN process.

Source retrieved 2026-09-03

Undermind is the research layer another agent orchestrates over MCP; it does not advertise several Undermind agents coordinating on one task.

Source retrieved 2026-09-03

Agents are composable steps inside a Play rather than peers negotiating a task; multi-step agents research, qualify and personalise.

Source retrieved 2026-09-03

Squads route one conversation across several specialised assistants with context-preserving handoffs.

Source retrieved 2026-09-03

Agents delegate sub-tasks to each other inside one workspace; the Plus plan lists multi-agent workflows as included.

Source retrieved 2026-09-03

The assistant fans work out to several background subagents at once, each with a researcher, builder or advisor role, though subagents cannot themselves spawn subagents.

Source retrieved 2026-09-27

Inbound Agent and TAM Agent coordinate through one Context Graph; the top plan adds unlimited AI Studio agents plus an Autopilot agent.

Source retrieved 2026-09-03

One agent spawns and coordinates children, which may run a different harness than the parent; orchestrations are one level deep and require approval before launch.

Source retrieved 2026-09-03

Chains run several research and extraction steps in sequence, each consuming the previous step's output, which is orchestration of stages rather than several agents coordinating concurrently on one task.

Source retrieved 2026-09-27

A parent agent can delegate to subagents with their own context windows, and separate threads run concurrently, optionally isolated in their own Git worktrees.

Source retrieved 2026-09-29

Several agents run at once in separate Git worktrees, and a single workflow can hand planning, implementation and review to different agents and models.

Source retrieved 2026-09-09

Admins can build many task-scoped custom agents alongside the channel AI agents, but Zendesk does not describe several agents coordinating on one task.

Swarm runs several agents at once as separate processes, each with its own model loop and session, and they share one working directory with no per-agent branch isolation.

Source retrieved 2026-09-28

Multi-agent orchestration is a named capability, with specialised research, analysis and action agents built on LangChain, CrewAI and custom orchestration layers.

Source retrieved 2026-09-25