Kalpana← all plans

§ enterprise

Self-hosted. Helm chart. Bring your own LLM. Talk to us.

Custom· Unlimited seats

§ what you get

Every plan ships with the same features.

Pricing scales with seats and support tier — the AI capability is identical across every plan. Unlimited seats on Enterprise.

Full access to your AI
Onboarding + business card
Up to 3 connectors
Local file access (Mac)
Business Mind
Chat with Claude (Sonnet + Haiku)
Unlimited connectors
Opus on demand
Background tasks
Priority support
Up to 10 seats
Role-based access
Shared business mind
Audit log + approvals
Slack + email digests
Up to 50 seats
SSO / SAML
SOC 2 evidence dashboard
Custom MCP bridges
White-label
Self-hosted (Helm)
Air-gapped deploys
BYO LLM keys
BAA / DPA
Custom SLA
Implementation support

§ what’s different at this tier

Why operators pick Enterprise.

01

Self-hosted via Helm — your VPC, your data, your perimeter.

02

Air-gapped deploys with offline model support (Ollama, vLLM).

03

BYO LLM keys — Anthropic, OpenAI, Bedrock, Vertex, Azure.

04

Custom SLA, BAA/DPA, named TAM, and implementation support.

§ a walk-through

From signup to shipped work in under 30 minutes.

  1. step 01

    Sign up

    Email + workspace name. We provision your tenant and email a secure download link in under 30 seconds.

  2. step 02

    Download the app

    Native macOS app, signed and notarized. Open it; it picks up your license automatically.

  3. step 03

    Onboarding tailors the AI

    Five questions, then it indexes your docs and connectors. The Business Card synthesizes who you are and how you work.

  4. step 04

    AI works on your business

    Drafts memos, triages inbox, runs research, builds tools. You review, approve, ship. It learns from every interaction.

§ what people use it for

Real jobs operators run on Enterprise.

case 01

Reviewing contracts and flagging redlines

case 02

SOC 2 evidence collection

case 03

Auto-generating spreadsheets from queries

case 04

Building internal tools without engineers

case 05

Compiling weekly metrics digests

case 06

Pulling product feedback from Slack + Linear

§ questions

Things people ask about Enterprise.

What infrastructure do we need?

Kubernetes 1.27+, Postgres 15+, Redis 7+, and an object store (S3 or compatible). 8 vCPU / 32GB RAM handles ~100 active users comfortably.

Can we run fully air-gapped?

Yes. Pair the Helm chart with Ollama or vLLM for inference. We ship a signed offline bundle with all dependencies pinned.

What's the implementation timeline?

Typical rollout: 2 weeks. Week 1 — install + connector wiring + IdP. Week 2 — Business Mind seeding + pilot cohort. Production by week 3.

Do you offer a custom SLA?

Yes. 99.9% uptime baseline; 99.99% available. Response-time tiers from 4h to 15min for P0 incidents. Pricing scales with the tier.

§ start now

Try Enterprise free for 14 days.

No card. No commitment. Download the app, finish onboarding, see if it fits your business.