LOOPmKit and CrewAI, side by side

What each one is

Where CrewAI leads

Rows where CrewAI is marked as found and LOOPmKit is not fully there.

• Self-hosting and licenseFramework is MIT. AMP Enterprise lists deployment in own VPC or own infrastructure. Source
• Hosted cloud offerFree tier with 50 executions per month on CrewAI cloud. Source
• Ready marketplace and community sizeAbout 60.5k GitHub stars. Source

What we did not find in the CrewAI docs

Rows where LOOPmKit has it built and the public pages we read for CrewAI do not say. Not found in the public docs we read is not the same as no.

• Agents trigger each other through records and eventsTriggers on a record, on another agent finishing, on a schedule or a webhook; also an agent as a tool of another. The live call of an agent as a tool awaited a key.
• Automatic routing to the cheapest suitable modelAuto mode picks the cheapest model that meets the tools, vision, context and quality needed, and a learned cascade starts on proven models.
• Hard budgets per run, agent and project that stop a runPer run, per agent per day and per project per day; a run over the cap is skipped, and parallel calls reserve budget first.
• Built-in savings: deduplication, response and prompt cacheThree savers on by default: dedupe by hash, an exact response cache and the provider prompt cache. No semantic cache.
• Trust levels: read, draft, actThe effective level is the lower of the action and the connector scope. Dynamic Gmail calls need act, since per-call approval is not supported yet.
• Evals and golden datasets10 kinds of evaluators, golden datasets, replay and comparison of versions and models. The prompt optimiser is not rolled out.
• Isolated code executionPer-job containers with a read-only root, dropped capabilities, limits and an egress allowlist. Component tests only; no run with a real vendor subscription yet.
• Company wizard with roles10 editable roles, up to 32, with budgets and colleague-call rules in one transaction. Checked in a browser with a scripted model. The goal of a company that runs itself is not reached.

Found in both

Rows marked as found for LOOPmKit and for CrewAI.

Visual canvas; Model providers, local models included; MCP as client and as server.

When to choose which

• Choose CrewAI when

  You want an MIT-licensed framework you can read and extend, a visual editor, a hosted cloud, and a community of about 60.5k GitHub stars.

• Choose LOOPmKit when

  You want a ready runtime with budgets per run, agent and project, an approval queue where the approver can edit arguments, and trust levels. The CrewAI pages we read list human-in-the-loop input but do not say more on these.

All rows and notes on the LOOPmKit side What is ready and what is not

Row by row

Each row shows LOOPmKit and CrewAI. Our notes come from our own feature inventory.

• Found
• Partly
• No
• ?Unknown: the pages read do not say

• Self-hosting and license

  LOOPmKit

  CrewAI

  LOOPmKit. A Docker Compose stack for one operator. The repository is private and has no license, so it is not open source and others cannot self-host it today.

  CrewAI. Framework is MIT. AMP Enterprise lists deployment in own VPC or own infrastructure. Source

• Visual canvas

  LOOPmKit

  CrewAI

  LOOPmKit. A canvas of agents and gates.

  CrewAI. Crew Studio, visual editor in the free tier. Source

• Agents trigger each other through records and events

  LOOPmKit

  ?CrewAI

  LOOPmKit. Triggers on a record, on another agent finishing, on a schedule or a webhook; also an agent as a tool of another. The live call of an agent as a tool awaited a key.

  CrewAI. No record-event agent chaining doc was checked; other forms of chaining may exist.

• Model providers, local models included

  LOOPmKit

  CrewAI

  LOOPmKit. 25 provider presets including Ollama, LM Studio and vLLM. The addresses were probed with invalid keys only.

  CrewAI. README: Ollama and LM Studio integration. Source

• Automatic routing to the cheapest suitable model

  LOOPmKit

  ?CrewAI

  LOOPmKit. Auto mode picks the cheapest model that meets the tools, vision, context and quality needed, and a learned cascade starts on proven models.

  CrewAI. Not found in the vendor docs read.

• Hard budgets per run, agent and project that stop a run

  LOOPmKit

  ?CrewAI

  LOOPmKit. Per run, per agent per day and per project per day; a run over the cap is skipped, and parallel calls reserve budget first.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Built-in savings: deduplication, response and prompt cache

  LOOPmKit

  ?CrewAI

  LOOPmKit. Three savers on by default: dedupe by hash, an exact response cache and the provider prompt cache. No semantic cache.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Approval queue with editing of arguments

  LOOPmKit

  CrewAI

  LOOPmKit. Drafts wait in a queue, the approver can edit the arguments, and disclosure is checked again after the edit.

  CrewAI. README: human-in-the-loop supported; pricing table lists human-in-the-loop input. Source

• Trust levels: read, draft, act

  LOOPmKit

  ?CrewAI

  LOOPmKit. The effective level is the lower of the action and the connector scope. Dynamic Gmail calls need act, since per-call approval is not supported yet.

  CrewAI. No read/draft/act ladder found in the docs read.

• Built-in persistent memory

  LOOPmKit

  CrewAI

  LOOPmKit. A separate memory service from the same author in the same stack; the run goes on if it is down. The "memory required" policy is not rolled out.

  CrewAI. README lists memory among agent capabilities without detail. Source

• Evals and golden datasets

  LOOPmKit

  ?CrewAI

  LOOPmKit. 10 kinds of evaluators, golden datasets, replay and comparison of versions and models. The prompt optimiser is not rolled out.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• MCP as client and as server

  LOOPmKit

  CrewAI

  LOOPmKit. As a client, any stdio or HTTP MCP server. As a server, 11 tools; built and not rolled out.

  CrewAI. README cites MCP/A2A support; AMP table lists MCP server export on both plans. Source

• Browser and desktop control

  LOOPmKit

  ?CrewAI

  LOOPmKit. Headless Chromium in the API image; a Mac bridge with manual setup; logged-in browser profiles are in progress.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Isolated code execution

  LOOPmKit

  ?CrewAI

  LOOPmKit. Per-job containers with a read-only root, dropped capabilities, limits and an egress allowlist. Component tests only; no run with a real vendor subscription yet.

  CrewAI. Not checked on the pages read.

• Company wizard with roles

  LOOPmKit

  ?CrewAI

  LOOPmKit. 10 editable roles, up to 32, with budgets and colleague-call rules in one transaction. Checked in a browser with a scripted model. The goal of a company that runs itself is not reached.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Number of connectors

  LOOPmKit

  ?CrewAI

  LOOPmKit. 61 presets and any MCP server or REST API. None is checked live one by one.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Telephony

  LOOPmKit

  ?CrewAI

  LOOPmKit. An inbound Twilio receptionist, tested against a simulated Twilio and off by default. No real call and no outbound calls.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

• Hosted cloud offer

  LOOPmKit

  CrewAI

  LOOPmKit. Self-hosting only is documented.

  CrewAI. Free tier with 50 executions per month on CrewAI cloud. Source

• Ready marketplace and community size

  LOOPmKit

  CrewAI

  LOOPmKit. A private repository and no public community. 23 templates inside the product.

  CrewAI. About 60.5k GitHub stars. Source

• SOC 2 or other enterprise certification

  LOOPmKit

  ?CrewAI

  LOOPmKit. No certification.

  CrewAI. Pricing page does not mention SOC 2.

• Customers

  LOOPmKit

  ?CrewAI

  LOOPmKit. No external customers and no revenue yet.

  CrewAI. Not stated on the vendor pages read; not assumed to be absent.

Checked on 4 November 2026. Notes on other products are quoted in English from the pages read.

Other comparisons

LOOPmKit vs n8nLOOPmKit vs DifyLOOPmKit vs LangGraphLOOPmKit vs Zapier Agents The full comparison of all nine products