LOOPmKit and LangGraph, side by side

What each one is

Where LangGraph leads

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

• Self-hosting and licenseLangGraph library is MIT. Self-hosted platform with control plane needs the Enterprise plan; standalone server needs a LangSmith license. Source
• Hosted cloud offerCloud in US or EU, Plus 39 USD per seat per month. Source
• Ready marketplace and community sizeAbout 44.5k GitHub stars. Source

What we did not find in the LangGraph docs

Rows where LOOPmKit has it built and the public pages we read for LangGraph 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.
• Model providers, local models included25 provider presets including Ollama, LM Studio and vLLM. The addresses were probed with invalid keys only.
• 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.
• 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 LangGraph.

Built-in persistent memory; Evals and golden datasets.

When to choose which

• Choose LangGraph when

  You write agents as code and want durable execution, long-term memory and evals in one ecosystem, with a hosted option in the US or the EU.

• Choose LOOPmKit when

  You prefer a canvas and a runtime that is already built, with budgets, an approval queue and trust levels in place. We did not find budgets, trust levels or model routing in the LangGraph pages we read.

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

Row by row

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

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

• Self-hosting and license

  LOOPmKit

  LangGraph

  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.

  LangGraph. LangGraph library is MIT. Self-hosted platform with control plane needs the Enterprise plan; standalone server needs a LangSmith license. Source

• Visual canvas

  LOOPmKit

  LangGraph

  LOOPmKit. A canvas of agents and gates.

  LangGraph. LangSmith Studio for visual prototyping; graphs themselves are defined in code. Source

• Agents trigger each other through records and events

  LOOPmKit

  ?LangGraph

  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.

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

• Model providers, local models included

  LOOPmKit

  ?LangGraph

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

  LangGraph. Library page read does not list providers.

• Automatic routing to the cheapest suitable model

  LOOPmKit

  ?LangGraph

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

  LangGraph. Not found in the vendor docs read.

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

  LOOPmKit

  ?LangGraph

  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.

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

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

  LOOPmKit

  ?LangGraph

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

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

• Approval queue with editing of arguments

  LOOPmKit

  LangGraph

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

  LangGraph. README: inspect and change agent state during execution; approval queue UI not verified. Source

• Trust levels: read, draft, act

  LOOPmKit

  ?LangGraph

  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.

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

• Built-in persistent memory

  LOOPmKit

  LangGraph

  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.

  LangGraph. README: long-term persistent memory across sessions and durable execution. Source

• Evals and golden datasets

  LOOPmKit

  LangGraph

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

  LangGraph. LangSmith: dataset collection, online and offline evals on all plans. Source

• MCP as client and as server

  LOOPmKit

  LangGraph

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

  LangGraph. Agent Server exposes each agent at /mcp; client side not verified here. Source

• Browser and desktop control

  LOOPmKit

  ?LangGraph

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

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

• Isolated code execution

  LOOPmKit

  ?LangGraph

  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.

  LangGraph. Not checked on the pages read.

• Company wizard with roles

  LOOPmKit

  ?LangGraph

  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.

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

• Number of connectors

  LOOPmKit

  ?LangGraph

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

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

• Telephony

  LOOPmKit

  ?LangGraph

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

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

• Hosted cloud offer

  LOOPmKit

  LangGraph

  LOOPmKit. Self-hosting only is documented.

  LangGraph. Cloud in US or EU, Plus 39 USD per seat per month. Source

• Ready marketplace and community size

  LOOPmKit

  LangGraph

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

  LangGraph. About 44.5k GitHub stars. Source

• SOC 2 or other enterprise certification

  LOOPmKit

  ?LangGraph

  LOOPmKit. No certification.

  LangGraph. Pricing page links a Trust Center, content not read.

• Customers

  LOOPmKit

  ?LangGraph

  LOOPmKit. No external customers and no revenue yet.

  LangGraph. 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 CrewAILOOPmKit vs DifyLOOPmKit vs Zapier Agents The full comparison of all nine products