Agentic Index

AI agent platforms that can be self hosted or run in a private VPC (2026)

Of the 548 agentic AI platforms in the Agentic Index, 275 document deployment inside the customer's own boundary in full, covering self hosted, on premises, air gapped and private VPC options. Only 165 also document bringing your own model or routing between providers. That is 30.1% of the field. This is a bar, not a leaderboard: a platform either documents both in its own public materials or it does not clear, and Partial evidence on either does not count.

The reason the bar has two parts rather than one is the finding underneath the list. Self hosting the software is not the same as containing the data. 110 of the 275 platforms you can run inside your own boundary do not document bring your own model. On those the runtime sits in your VPC and the model call still leaves it, so the prompt, the context and whatever the agent retrieved go to a third party anyway. A buyer who asks only the deployment question gets the wrong half of the answer about two times in five.

Which makes the scarcity ordering the second thing worth carrying out of this page. Of the four capabilities that bear on containment, model flexibility is the scarcest, not deployment. Across the same 548 platforms, full coverage runs observability 317, security and identity 320, deployment and residency 275, and model flexibility and routing 207. It is also the binding constraint on this bar: 110 of the 152 platforms one capability short fail on model flexibility alone. Separately, 42 platforms will take your model but will not run in your environment, which is the right answer for a team whose constraint is which model sees the data rather than where the software runs. Every platform that documents bringing your own model or routing between providers is named on its own page, counted from the model side rather than this one.

The bar, and how the 548 platforms score against it

Capability	What has to be documented	Full	Only blocker
Deployment and data residency	the platform documents running inside the customer's own boundary, self hosted, on premises, air gapped or in a private VPC, or documents where the data physically sits when it cannot	275 (50%)	42
Model flexibility and routing	the platform documents bringing your own model or key, or routing between providers, including local or self hosted models, rather than only calling one vendor hosted model	207 (38%)	110

Full means the vendor publishes evidence meeting the capability in its own public materials, under the Agentic Index verification standard. Only blocker counts platforms that document the other capability in full and fail on this one alone. One honest limit on the first row: deployment and data residency grades a spectrum, from customer controlled self hosting down to documented regional residency inside the vendor's own cloud. Read each platform's note for which end of that spectrum it clears on.

Clears the bar and scores 12.5 or higher of 14 overall

These 36 platforms document deployment inside your boundary and model flexibility in full, and also sit at the top of the Agentic Index coverage score across all 14 capabilities. Ordered by total coverage, ties broken alphabetically.

1. 1.Appian

   14.0 / 14 capabilities

   enterprise operations agents, agent builders

   Appian runs as managed SaaS across regions, in Appian Government Cloud inside an AWS GovCloud VPC with AI at feature parity, or self managed on Kubernetes. Each agent uses the model set in its model selection, defaulting to the latest from the configured provider.

2. 2.FLOWX.AI

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms, agent builders, agent infrastructure platforms

   Enterprise agentic platform whose documentation carries a named path for installing and configuring FlowX in the customer's own infrastructure, with every backend service containerized. Model providers are customer configured with the customer's own keys, so both halves of this bar sit in the docs rather than a sales deck.

3. 3.Gumloop

   14.0 / 14 capabilities

   agent builders

   Visual no code canvas for AI pipelines and agents that documents both halves of containment: deployment into the customer's own cloud as a VPC in the region of their choice with data remaining on infrastructure the customer controls, and model choice per agent across every major provider through Agent Preferences, with new models typically available within a day of public release.

4. 4.Mastra

   14.0 / 14 capabilities

   agent infrastructure platforms

   The Apache 2.0 framework builds and hosts agents anywhere, the licensed self hosted Enterprise edition keeps traces, prompts and outputs inside the customer's VPC, and Enterprise can run the Mastra Platform on premises or in the customer's own cloud. Each agent's model is a provider and model string in the model router, with the customer's own keys accepted.

5. 5.Microsoft

   14.0 / 14 capabilities

   enterprise operations agents

   Copilot Studio contains data by geography rather than by self hosting: data residency follows the environment's geography, with pages for data locations and for moving data across regions when generative features require it. The maker chooses each agent's primary model, an external model included, from published availability tables.

6. 6.ServiceNow

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms

   ServiceNow's containment answer is a named environment rather than self hosting: the Government Community Cloud under FedRAMP beside its standard cloud. An administrator picks the model provider for AI agents from Now LLM Service, AWS Claude, Azure OpenAI and Google Gemini.

7. 7.UiPath

   14.0 / 14 capabilities

   enterprise operations agents, multi-agent platforms, agent builders

   Agentic automation orchestrating agents, robots and people end to end. Deployable cloud, on premises or hybrid, with Automation Suite to run the full platform yourself, and Autopilot runs on the customer's cloud of choice or their own infrastructure while supporting GPT, Gemini, Claude and any compatible model.

8. 8.Pega

   13.5 / 14 capabilities

   enterprise operations agents, agent builders

   Infinity 26 runs on Pega Cloud or in the client's own managed cloud, with Pega Cloud for Government on AWS GovCloud and an EU service boundary controlling where European data lives. The GenAI layer is provider agnostic across Azure OpenAI, AWS Bedrock and Google Vertex, a choice left to the customer's architects.

9. 9.StackAI

   13.5 / 14 capabilities

   enterprise operations agents, agent builders

   No code enterprise agent platform with a dedicated On-Premise page, an enterprise CLI documented for deploying and managing StackAI on customer infrastructure, and an LLM hosting tree covering models in the customer's own Azure or AWS Bedrock account plus Local LLM for models on the customer's own hardware, governed through an LLM provider governance layer.

10. 10.Agno

    13.0 / 14 capabilities

    agent infrastructure platforms

    High performance agent runtime, formerly Phidata, Apache 2.0 licensed, which is the strongest possible containment position because the runtime is yours. Same evidence caveat as CrewAI: both rows cite the expansion evidence report rather than named documentation pages, a thinner basis than the enterprise platforms carry.

11. 11.AutoGPT

    13.0 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Self hosting is a right rather than a tier: the platform is open source and runs free via Docker Compose, with the customer's own model keys keeping data and inference on its own infrastructure, and Llama available for local model execution. The hosted cloud is the paid option.

12. 12.Base44

    13.0 / 14 capabilities

    agent builders

    Base44 does not self host. Enterprise workspaces choose US, EU or UK storage for app data on the vendor's monitored private infrastructure, which answers residency but not control of the machine.

13. 13.CrewAI

    13.0 / 14 capabilities

    multi-agent platforms

    Open source multi agent orchestration for collaborative crews: the MIT framework runs anywhere the customer runs Python, the enterprise platform deploys on CrewAI's cloud, a dedicated VPC or the customer's own infrastructure through CrewAI Factory, and each agent can take its own model from many providers.

14. 14.Databricks Agent Bricks

    13.0 / 14 capabilities

    multi-agent platforms

    Runs on AWS, Azure and Google Cloud with a published supported regions list and can be configured against the customer's own cloud account, while administrators connect external model providers on their own keys, alongside hosted open and frontier models, as Unity Catalog securables granted per user.

15. 15.Dataiku

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    Runs as managed Dataiku Cloud or installed on the customer's own infrastructure, with hybrid options through sales, and the LLM Mesh connects most major model providers plus locally running Hugging Face models that keep inference in house.

16. 16.HappyRobot

    13.0 / 14 capabilities

    enterprise operations agents

    AI workers for operations teams that document containment on both axes: the platform runs in HappyRobot's managed cloud, as a dedicated single tenant deployment with per customer encryption keys, in the customer's own VPC on AWS, GCP or Azure, or on premises, and workflows can run on OpenAI, Google Gemini, Anthropic Claude or models the customer hosts.

17. 17.n8n

    13.0 / 14 capabilities

    agent builders

    Open source workflow automation whose install documentation runs from one line setup, Docker and npm through Kubernetes and every major cloud, alongside managed n8n Cloud. Models attach to the AI Agent node as sub nodes, so different agents in one workflow can run different chat models, and a fully isolated configuration is documented for closed environments.

18. 18.Relevance AI

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    Hosted rather than self hosted, but an organization picks its region at signup from the US, EU or Australia, and each agent picks its model from several providers with a backup model from a different one; bring your own model keys is available on every plan.

19. 19.Replit Agent

    13.0 / 14 capabilities

    agent builders

    Full stack app builder and runtime whose Full here is about what you build, not where Replit runs: deployment coverage is the documented autoscale, scheduled, static and always on publishing types for customer apps, with intelligent model routing plus a manual model selector on the agent itself. Read it as containment for the apps it ships rather than a self hosted platform.

20. 20.Sema4.ai

    13.0 / 14 capabilities

    agent builders

    Enterprise agent platform from the Robocorp team whose agents run in the customer's own AWS, Azure, Google Cloud or Snowflake account, with the Enterprise Edition running entirely inside the customer's VPC, under a stated bring your own model approach naming OpenAI, Azure, Bedrock and Snowflake Cortex.

21. 21.Sim

    13.0 / 14 capabilities

    multi-agent platforms, agent builders

    Open source Apache 2.0 agent workspace. The sharpest single instance on this page: documented Docker Compose and Kubernetes self hosting including a fully offline setup with local models through Ollama, bring your own key at base provider pricing with no markup, and enterprise permission groups that restrict which providers are allowed.

22. 22.Tray.ai

    13.0 / 14 capabilities

    agent infrastructure platforms

    A Tray instance runs in a region the customer picks, US, EU or APAC, and an on premise agent with AWS PrivateLink reaches systems behind the firewall. Agents run on Tray's native model or on the customer's own OpenAI or AWS Bedrock keys, chosen per agent.

23. 23.Tungsten Automation

    13.0 / 14 capabilities

    enterprise operations agents, agent builders

    TotalAgility runs on the customer's own premises as well as in TotalAgility Cloud, and its Bring Your Own Model capability lets customers combine providers such as Azure OpenAI with their own custom LLMs.

24. 24.Warp

    13.0 / 14 capabilities

    multi-agent platforms (secondary lane membership, primary category Coding agent)

    Containment on both axes. Enterprise documents self hosted cloud agents on the customer's own infrastructure with inference routed through the customer's own cloud, Factories runs in Warp's cloud or fully self hosted inside the customer's VPC, and a model picker spans Claude, GPT, Gemini, GLM and Kimi models, with different models on different tasks.

25. 25.Writer

    13.0 / 14 capabilities

    enterprise operations agents, agent infrastructure platforms

    Enterprise platform on its own Palmyra model family whose enterprise deployment documentation names self hosting and full data residency as available options, meeting the deployment bar three ways where one would do. The model half clears on the vendor's own FAQ answering who chooses the model.

26. 26.Akka

    12.5 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Runs self managed on the customer's own VMs, containers or Kubernetes, in the customer's VPC through Akka Automated Operations, or in Akka's serverless cloud, and each agent can take its own model from nine built in providers or a custom one, including local models through Ollama.

27. 27.Boomi

    12.5 / 14 capabilities

    enterprise operations agents, multi-agent platforms, agent builders

    Integration platform with an agent control tower over enterprise systems. Runs on a flexible multi cloud autonomous runtime with on premise and multi region deployment options for residency, and is model agnostic with MCP support across Amazon Bedrock, Snowflake Cortex and OpenAI, with roadmap support for custom large and small models.

28. 28.Fabrix.ai

    12.5 / 14 capabilities

    enterprise operations agents, agent builders

    Deployable on premises or in the customer's private cloud with the model layer included, so open source models such as Llama and Mistral can run inside the customer's infrastructure, beside OpenAI and Anthropic as cloud choices the customer selects.

29. 29.Harvey

    12.5 / 14 capabilities

    enterprise operations agents

    Enterprise legal platform, the one vertical product in this tier. It clears deployment on documented regional hosting rather than self hosting, with the customer keeping everything in region across EU or Switzerland, US, or Australia and separate regional application hosts, and clears model flexibility on admin controlled rollout of Anthropic and Google models beside OpenAI with an opt in model selector.

30. 30.IBM watsonx Orchestrate

    12.5 / 14 capabilities

    enterprise operations agents

    Runs as a managed service on IBM Cloud or AWS or in the customer's own on premises environment, with documented air gapped installation. Models from several providers are imported and assigned per agent, coordinated by policies for load balancing and fallback.

31. 31.Kestra

    12.5 / 14 capabilities

    multi-agent platforms, agent infrastructure platforms

    Apache 2.0 open source that self hosts on the customer's infrastructure, cloud or Kubernetes, with the Enterprise Edition on cloud, on premises or air gapped. The model is chosen on each AI task through its provider property, from Anthropic and OpenAI to Bedrock, Gemini, Mistral and local Ollama.

32. 32.Latenode

    12.5 / 14 capabilities

    agent builders, agent infrastructure platforms

    Low code agent and automation platform documenting self hosting for teams prioritizing data security, with full control over infrastructure and data across on premise and cloud environments, and one subscription covering GPT, Claude, Gemini, Deepseek, LLaMA and a long tail of further models without separate provider accounts.

33. 33.OutSystems

    12.5 / 14 capabilities

    multi-agent platforms, agent builders

    Enterprise low code incumbent whose ODC self hosted deployment, published March 2026, runs apps, data and agents in the customer's private cloud, public cloud or on premises Kubernetes while OutSystems manages platform services, and whose Agent Workbench connects a model once for reuse across agents, from Bedrock, Azure OpenAI and Anthropic through custom models on Vertex AI.

34. 34.Salesforce

    12.5 / 14 capabilities

    enterprise operations agents

    CRM platform whose Agentforce layer runs agents across sales, service and marketing, with the Atlas Reasoning Engine supporting OpenAI, Anthropic and Google Gemini. It clears this bar on data residency and the trust layer, not on self hosting, and its own filed evidence says so. Read this note before treating it as an answer to this question.

35. 35.SimplAI

    12.5 / 14 capabilities

    multi-agent platforms, agent builders, agent infrastructure platforms

    Datafuse Technology Inc builds SimplAI, an enterprise agentic AI operating system spanning SimplAI cloud, customer cloud, private cloud, VPC, on premises and air gapped with no architectural change between them, plus customer defined retention and deletion per agent. One caveat a buyer should carry: bring your own models and fine tuning are documented but gated to the enterprise tier.

36. 36.Torq

    12.5 / 14 capabilities

    enterprise operations agents (secondary lane membership, primary category Security / SOC agent)

    Security hyperautomation platform whose self hosted step runners run workflow steps inside the customer's own environment, including on EKS and AKS, while workspaces sit in the United States, the European Union or Japan. Customers bring their own model subscription across OpenAI, Azure OpenAI, Vertex AI, Anthropic and Bedrock and pick the model per agent and per task. Qualifies here on secondary lane membership as an enterprise operations platform, which a buyer scanning this list should know.

The remaining 129 platforms that clear the bar

Every one of these documents deployment inside the customer's boundary and model flexibility in full. They score below 12.5 of 14 on total coverage, which says something about breadth across the whole taxonomy rather than about how well contained they are. Several of the strongest containment stories in the index are narrow open source runtimes that score low precisely because they do one thing.

Common questions

Which AI agent platforms can be self hosted or run in a private VPC?

275 of 548 agentic AI platforms in the Agentic Index document deployment inside the customer's own boundary in full, covering self hosted, on premises, air gapped and private VPC options. Only 165 of those also document bring your own model or routing between providers in full, which is the bar used on this page, because a runtime inside your boundary that can only call one vendor hosted model has not contained your data. That is 30.1% of the field. Graded from public evidence only.

Is self hosting an AI agent platform the same as keeping your data private?

No, and this is the gap the page exists to show. 110 of the 275 platforms documenting deployment in your own boundary do not document bring your own model. On those platforms the software runs inside your perimeter and the model call still leaves it, so the prompt, the context and whatever the agent retrieved go to a third party anyway. Self hosting the runtime and containing the data are two different purchases.

Which is harder to find, self hosting or model flexibility?

Model flexibility. Across the same 548 platforms, full coverage runs observability 317, security and identity 320, deployment and residency 275, and model flexibility and routing 207. It is the scarcest of the four, and it is the binding constraint here: 110 of the 152 platforms one capability short of the bar fail on model flexibility alone.

Can you point an AI agent platform at your own model without self hosting it?

On 42 platforms, yes. They document bringing your own model or routing between providers while running in the vendor's cloud rather than yours. That is a real option for teams whose constraint is which model sees the data rather than where the software runs, and it is a different shape of answer from the 165 platforms that document both.

Does every platform on this list actually let you self host?

Two do not, and the page names them. The deployment and data residency capability grades a spectrum, from customer controlled self hosting down to documented regional residency inside the vendor's own cloud. Two cleared platforms have filed evidence stating in terms that customer self hosting is not the model. They stay in the list because the rule is mechanical and applied to every vendor equally, and the disclosure sits beside their entries so nobody reads them as self hostable.

Is this ranking paid or sponsored?

No. No vendor pays for placement, no vendor has reviewed this page, and every grade comes from the vendor's own public materials under the Agentic Index verification standard. 956 vendors are graded against the same 14 capabilities. Data last verified October 3, 2026.

Method: membership is the same 548 platform pool used by the best agentic AI platforms in 2026, drawn from 956 researched vendors. The editorial pages over this pool are one method with different bars, not several opinions. Every grade comes from the vendor's own public materials under the Agentic Index verification standard. No vendor pays for placement and no vendor has reviewed this page. Data last verified October 3, 2026. How this evidence is graded

Related: enterprise security and compliance platforms, platforms that require human approval, agent observability platforms, how every vendor scores on deployment and data residency, platforms that let you bring your own model or route between providers, how every vendor scores on model flexibility, agent infrastructure platforms ranked on the production contract, compare platforms side by side.