Pricing

Start free. Scale when you're ready.

Unlimited usage on hardware you own — no token meters, no rate limits.

Free

One user, one GPU. The full model catalog on your own hardware — free forever.

Full engine + entire model catalog
Local chat app — nothing leaves your box
Gmail + telephony (SMS & voice)
API access → Starter
318 connectors → Pro

Download free

Starter

$19/user/mo

For builders. Program against your model and ship products on it.

Everything in Free, plus
OpenAI + Anthropic-compatible /v1 API
Bring your own model — load any Hugging Face repo
Bare Metal Code — local coding agent + editor
Presence — local avatar agent (Preview)
Curated dev connectors + custom MCP servers

Pro

$39/user/mo

Everything connected. Wire the model into your whole stack.

Everything in Starter, plus
All 318 connectors & integrations
Team sharing — each seat shares with 4
Priority email support

Enterprise

Custom

Scale across your org with a fleet and governance.

Everything in Pro, plus
Multi-GPU fleet — tensor parallelism over your network
SSO / SAML + SCIM + RBAC
Audit logging & multi-site fleet management
Model-access policy & governance

Talk to us →

Custom

Let's talk

White-glove. We deploy it; you own it.

Everything in Enterprise, plus
Managed appliance — we provision & ship a box you own
Air-gapped deployment & custom integration
Named account team + support SLA
Hands-on onboarding

Talk to us →

No rate limits on any tier — you own the hardware, run it as hard as you like. Trials unlock the full tier; when one ends, configured connectors pause read-only rather than break.