Pricing·Free app · paid inference
$0/app

The app is free.
Inference is optional.

BYOK included · Plus from $25/mo

Download Continuum and bring your own Claude, Codex, or other provider keys - no Continuum fee. When you want Continuum-hosted inference, personal API keys, and a weekly budget, pick Plus, Max, or Ultra.

01 · Personal or organization

Buy for yourself, or buy for a team. The app and the models are the same either way; an organization adds model policy, weekly caps, and one bill.

Personal $0/mo to start

one person · the whole app, free · individual setup

Code on every platform: Mac, iPhone, web, Windows, and Linux
Your own provider keys, with no Continuum fee
Hosted inference from $25 a month whenever you want it
Personal Continuum API keys on any paid plan
Usage analytics for the work you run locally

Free $0 · Plus $25 · Max $100 / $200 · Ultra $500

Organizations from $25/member/mo

teams and enterprises · billed by live member count

Team $25 · $100 · $200 · $500
Plus, Max, or Ultra seats on one subscription
Model policy by team and person
Weekly caps and approvals
Usage and spend, one view
Your keys or ours
Enterprise let's talk
Everything in Team
Dedicated account management
Highest priority support
On-prem deployment
Custom inference, including GPU installation and management
02 · Personal plans

Hosted bundles buy Continuum-routed inference with a weekly spend budget. The Mac, web, and iPhone apps stay free either way.

Free $0/mo

the whole app · bring your own keys

Code and Usage across Mac, web, and iPhone
All providers via your own subscriptions or API keys
OSS / free models without a Continuum bill
Usage analytics for the work you run locally
Plus $25/mo

hosted inference included

Everything in Free
$25 of hosted usage every week
Continuum-hosted frontier models
Personal Continuum API keys
No BYOK setup required
Start 7-day free trial of Continuum Plus

7 days free, then $25/month, cancel anytime. Card required. One trial per account.

Max 200 $200/mo

8× the Plus allowance

Everything in Plus
8× the Plus weekly hosted usage ($200 a week)
Headroom for multi-agent weeks
Still bring your own keys anytime
Ultra $500/mo

40× the Plus allowance

Everything in Max 200
40× the Plus weekly hosted usage ($1,000 a week)
Twice the hosted usage per dollar of any other plan.
Built for heavy multi-agent weeks
Still bring your own keys anytime
02b · Compare at a glance
Feature Free Plus Max 100 Max 200 Ultra
Price / month$0$25$100$200$500
Full Continuum app
Bring your own keys
OSS / free models
Hosted inferenceno
Weekly hosted usageno$25 of hosted usage / week4× Plus ($100/wk)8× Plus ($200/wk)40× Plus ($1,000/wk)
Personal API keysno
03 · Model pricing

The rate, model by model.

Standard hosted rates are shown per million tokens. When a model has a promotion, the same weekly allowance buys 1.5×, 2×, or 3× more usage for the dates shown.

Live rates · allowance equivalent · UTC dates

ModelInput / 1MOutput / 1MPromotion
Loading current model rates…

Promotions apply only to Continuum-hosted inference. The effective rate is the allowance equivalent: your plan consumes less allowance for the same model usage during the active window.

04 · Two ways to run models

Bring your own keys and Continuum never sits in your token path - you pay providers exactly what you already pay them. Or subscribe for Continuum-hosted inference when you want a single budget, personal API keys, and no per-vendor setup.

App access is free either way.

bring your own keys$0 Continuum

providers bill you directly · no Continuum markup

hosted inferencefrom $25/mo

weekly budget · Continuum routes · personal API keys

Each is a five-minute setup: one key, one base URL. Personal API keys start on Plus.

05 · What you actually pay

On Free, Continuum is a $0 line. On Plus, Max, or Ultra, you pay one inference plan and keep using BYOK whenever you want.

A typical month with your own subscriptions still goes straight to Anthropic, OpenAI, and the rest. Continuum only appears on the receipt when you choose a hosted plan.

Running a team? One subscription, billed by your live member count. See Continuum for organizations →

Optional hosted subscriptions and account use are covered by the Continuum Terms of Use.

receipt · a month with Continuum#0001
Claude Max$200.00 → Anthropic
ChatGPT / Codex$20.00 → OpenAI
Cursor Pro$20.00 → Cursor
Continuum app$0.00
Plus hosted (optional)$25.00 /mo
Continuum app$0.00
06 · Questions

The honest fine print.

Short answers here, longer ones in the docs.

Yes. Continuum's Code and Usage surfaces are free with your own provider keys. You only pay Continuum if you want hosted inference (Plus, Max, or Ultra).

Continuum-hosted inference with a weekly budget, personal API keys, and no per-vendor BYOK setup. Plus is $25/mo, Max 100 is $100/mo, Max 200 is $200/mo, and Ultra is $500/mo. Each step up raises the weekly hosted allowance: Max 100 is 4× Plus, Max 200 is 8× Plus, and Ultra is 40× Plus.

Yes. Hosted inference is additive. You can keep routing sessions through your Claude, Codex, or other provider accounts anytime.

Sign in to sync devices and manage an optional inference plan. The app still runs local-first - your prompts, code, and logs stay on your machine.

Continuum Plus offers a 7-day free trial through Stripe on web, Mac, iOS, and Expo. A card is required. After 7 days, Plus is $25/month unless you cancel. One trial per account. The Continuum app remains free with your own provider keys.

Individuals download and start free. Teams buy seats on Plus, Max, or Ultra as one subscription billed by live member count, which adds model policy by team and person, weekly caps and approvals, and usage and spend in one view. Enterprise engagements cover on-prem deployment and custom inference. Book a demo if you would rather talk it through first.

07 · Begin

Download it.
Start in five minutes.

The app is free. Bring the subscriptions you already pay for, or add Plus later for Continuum-hosted inference. Running a team? Buy seats on the org platform.

requires an apple silicon mac · iphone & watch optional