Self-hosted: a census of 145 listings

Self-hosting is the only mode: the daemon, proxy and plugin server all run on your machine over loopback ports, and the vendor states work never leaves the box.

Source retrieved 2026-10-08

Client-owned deployment is offered but explicitly conditional on the architecture permitting it, so a buyer cannot assume their own infrastructure on every engagement.

Source retrieved 2026-10-08

Everything runs wherever you can run Node, including your own servers, edge runtimes and desktop hosts, because the runtime is process-agnostic rather than a hosted service.

Source retrieved 2026-10-08

The runtime and database are yours to host anywhere a container runs; only the control plane UI is hosted by default, and a self-hosted control plane is an Enterprise item.

Source retrieved 2026-10-09

Standard deployment is Aide's Azure US East 2 tenancy, but the vendor says it can provision into public or private clouds where residency demands it; that is vendor-provisioned, not customer-run, and no tier or terms are given.

Source retrieved 2026-09-23

On-premises deployment is offered only through a sales conversation on the custom tier; no self-hosted option exists on any published self-serve plan.

Source retrieved 2026-10-03

Aider is installed on your own machine with pip, uv, pipx or Docker and edits files in your local repo, so there is no vendor runtime to depend on.

Source retrieved 2026-09-30

The vendor claims VPC, hybrid and on-premises deployment alongside the public clouds, but publishes no installation documentation; the only on-prem component documented is the Remote Executor connector.

Source retrieved 2026-09-24

The binary and its sandbox run on your own machine and requests can be pointed at a Gemini-compatible endpoint or your own Cloud project behind VPC-SC, but the models themselves stay Google-hosted.

Source retrieved 2026-09-10

The system Arsum builds runs in an environment the client owns, with code and account ownership settled at handover, but there is no Arsum platform to self-host.

Source retrieved 2026-10-03

Agents can execute on your own machines via daemons, but inference and the control plane stay with Augment.

Source retrieved 2026-09-02

Everything runs on your own machines: agents execute in your process, and a gRPC host plus worker runtimes let you distribute them across your own infrastructure.

Source retrieved 2026-09-19

Running models on the client's own hardware is a delivered pattern with a published case study behind it, and a company blog argues for deploying the AI inside the customer's own cloud.

Source retrieved 2026-09-11

VPC, on-premise and air-gapped deployments exist but the pricing table gates them to Enterprise contracts only.

Source retrieved 2026-09-11

Self-hosting exists only as an Enterprise BYOK deployment into your own AWS or Azure tenant; Free, Pro and Teams are hosted by Bolt only.

Source retrieved 2026-09-11

On-premise or VPC model deployment is sales-gated; separately, agent reasoning code can run on your own servers with Cartesia connecting to your URL.

Source retrieved 2026-09-02

Self-hosting is the default: your process spawns a claude CLI subprocess with your working directory and disk.

Source retrieved 2026-09-02

The CLI executes on your own machine, and cloud sessions can run on your runners in public beta on Team and Enterprise; inference always goes to Anthropic or a cloud provider.

Source retrieved 2026-09-02

The client executes on your own machine by default, and Enterprise is documented for VPC, on-premises and air-gapped deployment with inference going straight to your providers.

Source retrieved 2026-09-12

A container image runs the review agent inside your own network, but only for Enterprise customers at 500 seats or more, arranged through sales during onboarding.

Source retrieved 2026-09-15

The agent binary runs on your machine or runner; inference is OpenAI-hosted unless you route it through your own AWS Bedrock account.

Source retrieved 2026-09-02

The open-source framework installs and runs on your own machines on any tier, while self-hosting the managed platform (VPC, own infrastructure, Factory) is an Enterprise option.

Source retrieved 2026-09-13

Only tool execution moves to your machines: Cursor still runs the agent loop, inference and planning, and the Self-Hosted Pool needs Enterprise. The Enterprise FAQ says there is no on-premises deployment.

Source retrieved 2026-09-02

Decagon claims headless deployment inside your own infrastructure, but publishes no on-premise or self-hosted edition and runs its platform on its own cloud subprocessors.

Source retrieved 2026-09-14

Self-hosted containers are an Enterprise-only path. Deepgram's own self-hosting pages enumerate speech-to-text, text-to-speech and audio intelligence, not voice-agent orchestration, while the Voice Agent product page lists self-hosted as a deployment option. Dedicated single-tenant hosting does explicitly cover agent workloads.

Source retrieved 2026-09-02

Outposts runs session execution on machines you operate, but the agent loop stays in Cognition's cloud.

Source retrieved 2026-09-02

Enterprise-only, and the deployment docs describe Cognition hosting in a customer-isolated single-tenant VPC reached over AWS PrivateLink rather than customer-run software.

Source retrieved 2026-09-03

Community Edition runs anywhere Docker runs; Enterprise adds Kubernetes/Helm, VPC and air-gapped deployment.

Source retrieved 2026-09-02

The isolation option Dust sells is a dedicated single-tenant deployment it runs on Enterprise; the MIT-licensed source exists but no self-hosting guide is published.

Source retrieved 2026-09-09

The full agent runtime can run in your own VPC or hardware on AWS or GCP, but the documentation is NDA-gated and sold through the account team; Azure is unreleased.

Source retrieved 2026-09-02

Only the Enterprise column mentions custom deployments such as single-tenancy, and Elicit never documents that as running on customer infrastructure; every self-serve plan is vendor-hosted only.

Source retrieved 2026-09-14

The agent always runs on your machine, and a dedicated airgap build exists, but airgap builds are Enterprise-only and not publicly distributed.

Source retrieved 2026-09-02

Docker Compose self-hosting runs the core scrape, crawl, map and search routes, but agent, browser, interact and enterprise controls stay Cloud-only.

Source retrieved 2026-09-24

Claw can run on your own Mac or Windows machine and GenTeam agents can run on a computer you connect, but the workspace, accounts and everything else stay on Genspark's cloud.

Source retrieved 2026-09-15

Customer Hosted runs Glean inside your own AWS or GCP account, but as a Glean-operated managed service you may not deploy or patch yourself.

Source retrieved 2026-09-02

Self-hosting is a documented first-class path: package the agent as a container and run it anywhere, including locally in Docker or Podman with no Google Cloud connection.

Source retrieved 2026-09-15

Self-hosting is the only deployment model: the agent, its sessions and its logs live on your machine, and goose serve can be run on your own VM with TLS and a shared secret.

Source retrieved 2026-09-16

Self-hosting is the only supported deployment: pip package, FastAPI server, docker-compose or Linux host, all on infrastructure you own.

Source retrieved 2026-09-16

Virtual Private Cloud deployment is an optional Enterprise line item only, and Gumloop operates the stack inside your cloud rather than handing you software to run.

Source retrieved 2026-09-30

Enforcement already runs entirely on your own machine, but a self-hosted or VPC control plane is only an Enterprise review item.

Source retrieved 2026-09-29

Self-hosting is a separately purchased line, IrisAgent Private, installed in your data centre or your own AWS, Azure or GCP account; the standard Free and Standard plans are vendor-hosted SaaS.

Source retrieved 2026-09-24

Self-hosting is the only deployment there is: you run the server on your workstation, a dev box or your own server, and there is no vendor cloud in the path at all.

Source retrieved 2026-09-17

The agent binary always runs on your machine and models can be local, but running the surrounding JetBrains AI service on your own infrastructure is ticked for AI Enterprise alone.

Source retrieved 2026-09-18

The MIT clients can be forked and self-hosted and inference can be fully local, but Gateway, Cloud Agents and team administration are Kilo-hosted.

Source retrieved 2026-09-02

The open source frameworks run anywhere, but self-hosting or hybrid-hosting the LangSmith platform on Kubernetes is an Enterprise-plan option; Developer and Plus are cloud-only.

Source retrieved 2026-09-18

Self-hosting is the primary delivery model: a Docker image, a pip-installable package and deployment guides for major clouds, all running on infrastructure you control.

Source retrieved 2026-09-25

The library is MIT-licensed and installed with pip or npm, so it runs wherever you install it; self-hosting the managed Agent Server platform is a separate Enterprise hosting option.

Source retrieved 2026-09-02

One comparison page offers optional on-prem infrastructure for sensitive environments, while the security page says data never leaves Leaping AI's own controlled environment; no self-hosting documentation, requirements or tier is published either way.

Source retrieved 2026-09-09

Local runtime keeps all state on-device with no account, or run the App Server via Docker, Railway or Fly.io.

Source retrieved 2026-09-02

Agents are containers you can run on your own Kubernetes; the LiveKit media server and SIP service are separately self-hostable. Self-hosted agents still meter WebRTC participant minutes if they connect to LiveKit Cloud.

Source retrieved 2026-09-02

Full platform deploys into your own Kubernetes/VPC, but only on Enterprise plans.

Source retrieved 2026-09-02

Four postures are sold, up to on-prem Kubernetes and a fully air-gapped build, and the pricing page attaches published floors to each rather than leaving them to sales.

Source retrieved 2026-09-09

mastra build emits a self-contained Node server that runs on any Node/Bun/Deno host; Enterprise Edition can be deployed on-prem or in your own cloud.

Source retrieved 2026-09-02

Self-hosting is one of the two documented hosting models: the agent runs in your own web service, container or runtime and you keep routing, identity, storage and scaling.

Source retrieved 2026-09-19

Self-hosting the platform is listed only under the Business plan's flexible deployment bullets; on Free and Individual you can self-host models but not MindStudio itself.

Source retrieved 2026-09-20

Docker, npm or Helm on your own boxes, air-gapped if required; the free Community edition needs no licence key, and Business or Enterprise features unlock with one.

Source retrieved 2026-09-02

Everything runs on the buyer's own machine by design, with canvases never copied to a vendor server and sharing streaming peer to peer.

Source retrieved 2026-10-08

Self-hosted delivery is demonstrated in at least one shipped engagement and offered for privacy-sensitive models, but it is scoped per project rather than a standing deployment option.

Source retrieved 2026-09-26

The framework and its local runtime tools execute entirely in your own process or environment; only model inference and optional trace export leave your infrastructure.

Source retrieved 2026-09-20

Self-hosting is the only deployment model: the Gateway runs on your macOS, Windows or Linux machine, a VPS, a container or Kubernetes, with no vendor-operated tier.

Source retrieved 2026-09-21

The agent runs on your own machine or server and can be pointed exclusively at an internal AI gateway with other providers disabled.

Source retrieved 2026-09-21

Two self-host routes: the free open-source stack on your own machine, and Enterprise deployed into your own Kubernetes VPC.

Source retrieved 2026-09-21

Only the quoted Enterprise tier offers infrastructure you operate yourself; the three published plans are hosted by the vendor with no self-hosting option.

Source retrieved 2026-10-03

There is no vendor runtime to depend on: the package installs into your own Python environment and the web app it produces runs under any ASGI server you operate.

Source retrieved 2026-09-21

'In VPC Deployment' is checked only in the Enterprise column: the warehouse stays the system of record and data need not leave the customer VPC, but compute runs in Rox's VPC, so this is not a customer-run install.

Source retrieved 2026-09-03

A dedicated cloud deployment is offered only on the enterprise-grade custom build; the entry packages describe hosting as quoted separately and nothing documents running it on your own hardware.

Source retrieved 2026-09-28

Docker Compose and Kubernetes deployments are documented and free; the pricing table nevertheless lists 'Self Hosting' only under Enterprise.

Source retrieved 2026-09-03

Your half (webhook server and LLM) runs anywhere you like, but the telephony, transcription and speech synthesis run on sipgate's own EU infrastructure with no self-hosted option documented.

Source retrieved 2026-09-26

The MCP server is hosted by the vendor and the setup page says nothing is installed locally, yet the homepage describes the client launching an npx package locally; the two pages disagree.

Source retrieved 2026-10-03

On-premise and VPC deployment exist but sit in the Enterprise column of the plan table; the free tier is multi-tenant cloud only, and deployment needs a licensed CLI.

Source retrieved 2026-09-09

Self-hosting is the Community edition's only deployment mode and remains an option on Enterprise, on your own hardware or in Tale's cloud.

Source retrieved 2026-10-03

Workflow and trading builds can be deployed onto the customer's own VPS or infrastructure, but the Agent-OS harness that runs the autonomous agents stays on TecAdRise's side.

Source retrieved 2026-09-12

Cloud-hosted by default; VPC deployment is an Enterprise item and on-premises needs a separate written agreement.

Source retrieved 2026-09-03

Self-hosting is real but doubly bounded: Agent Builder is named as included in Automation Suite on AKS, EKS or OpenShift including air-gapped, while the Automation Suite on Linux flavour has agentic AI only planned, and the Basic plan carries an em dash for deploying in your own environment.

Source retrieved 2026-09-03

Self-serve is hosted only; on-premise is offered to large enterprises, and the orchestration layer otherwise never leaves Vapi.

Source retrieved 2026-09-03

The runtime can run on your Mac, a GCP VM in your own project, or any Linux host you can SSH into, with no platform or compute fees.

Source retrieved 2026-09-27

Enterprise only, and execution plane only: agents run on your machines while orchestration, session management and inference still route through Warp's backend.

Source retrieved 2026-09-03

Everything but the chat itself runs on your own machine: files are created and commands executed locally by the installed CLI.

Source retrieved 2026-09-21

Models and edit prediction can run on hardware you control and the editor builds from source, but Zed's hosted AI, collaboration and org dashboard remain Zed-operated services.

Source retrieved 2026-09-29

Zenflow can be installed on a VPS you own, and a full private deployment running orchestration and inference inside your network is listed as an Enterprise-plan feature.

Source retrieved 2026-09-09

Everything runs on your own machine: one binary, no vendor service in the path, with all configuration, credentials and transcripts stored in a local state directory.

Source retrieved 2026-09-28

Systems they build can be deployed into the client's own environment, including on-premise and VPC-hosted options aimed at regulated healthcare and finance buyers.

Source retrieved 2026-09-25