Boski vs Martin vs Noah AI

The same 33-row checklist and the same rules. Every cell shows vetted official evidence or “Not checked.”

Visit Boski Visit Martin Visit Noah AI Boski alternatives Martin alternatives Noah AI alternatives

So which one?

Notis aside, Martin has the highest checked-evidence score: 41% across 33 vetted rows (all 33). Boski: 39% across 31 vetted rows; 2 rows are not checked yet and Noah AI: 34% across 32 vetted rows; 1 row is not checked yet.

They are not really after the same job, though: Martin for remembering context between conversations, Boski for scheduled and event-triggered work and Noah AI for reaching it from the apps you already message in.

Nothing here publishes anything on Bring your own API key, Open source and Run it on your own server.

Only Notis publishes anything on Telegram, Runs on your own Claude plan, Runs on your own ChatGPT plan and 4 more rows.

Not sure which of these is yours?Find your match in two minutesFree, and no email.

The whole grid, row by row

Scroll sideways to compare every product. Feature names stay pinned.

Capability comparison between Notis and the selected assistants
Capability	View pricing	BoskiVisit site	MartinVisit site	Noah AIVisit site
Where you can use it
WhatsApp
iMessage			—Not offered
Telegram		—	—Not offered	—Not in FAQ or terms channel lists
Slack		—	Vendor Slack app, DM or @mention
Email			Email him and he replies	CC or write to Noah's own address
SMS or a phone number	PartialInbound calls only, no SMS	—	Reachable via SMS or phone call
Voice messages	Voice notes + live calls
Web or desktop app	Web and desktop app		Web app only, no downloadable client	Partial
Its own iPhone or Android app	—	Partial	iOS only	—"No app to download"; no store listing
Bring your own AI
Runs on your own Claude plan		—	—Not offered	—No sign-in with a Claude plan
Runs on your own ChatGPT plan		—	—	—No sign-in with a ChatGPT plan
Bring your own API key	—	—	—Not offered	—Flat $49/month, model usage included
Works inside your own coding agent		—	—No MCP server, SDK, or public API	—No MCP server or developer docs
You choose which model runs your work		Not checked	—	—One undisclosed third-party LLM stack
What it can do
Ready-made app integrations	1,000+ apps	—	Partial3 pages, ~6 named apps	Partial9 named in FAQ; full list login-gated
Creates and edits documents			Partial	—Docs/Drive are referenced, not authored
Save your own reusable instructions		—		—No saved instruction sets documented
Browses the web for you		Not checked	—Search only, no page interaction	—No web browsing; errands go by phone
Posts to social media for you			—Not offered	—
Joins and records your meetings	—	—	—Not offered	Not checked
Memory
Remembers you between conversations			Pro only: "Enhanced long term memory"
Reads your existing files and apps	Notion, Gmail, Drive, Granola, web
Runs without you
Runs on a schedule or a trigger			Pro-only Background Tasks	Partial
Reminds and nudges you
Computers it can use
Gives it a cloud computer of its own	1h/mo (Pro) to uncapped (Ultra)	—	—Not offered	—
Controls your own computer	Shell and apps, via the desktop app	—	—Not offered	—Nothing runs on your machine
Connects to MCP servers on your computer		—	—Not offered	—No local agent; MCP not mentioned
Trust & control
Open source	—	—	—Closed SaaS	—Closed; proprietary IP clause
Run it on your own server	—	—	—Hosted subscriptions only	—Hosted only via heynoah.io
Says it will not train on your data	Covers Notis and its LLM providers	Partial	Partial	—Not stated (checked 2026-09-02)
Published security certification	PartialSub-processors certified; Notis is not	—	—	—
Plans & support
Free plan does the product's main job	Partial	—	—Trial only, no ongoing free plan	—Paid only: $49/month, invitation-gated
Shared team workspace		—	—Single-person subscriptions only	Partial
Price	FREEPRO$13/moyr (rate on a yearly commitment)+2 more	Tiers not collected	Basic$21/moyr (rate on a yearly commitment)Pro$30/moyr (rate on a yearly commitment)	Early adopter$49/mo

IncludedPartial— Not documentedNot checked

The catch with each one

Numbered by support score across vetted capability rows, with incomplete evidence shown on every card. Unknown rows never count as a no, and this is not a score against your answers.

1. Notis

Personal AI assistant

Notis is a personal AI assistant you message from WhatsApp, iMessage, Telegram, Slack or email, and reach from your own coding agent through its MCP server.

Best for: running long jobs on a machine of its own

Documented strengths

Where you can use it: WhatsApp, iMessage, Telegram, Slack, Email, Voice messages and Web or desktop app
What it can do: Ready-made app integrations, Creates and edits documents, Save your own reusable instructions, Browses the web for you and Posts to social media for you
Bring your own AI: Runs on your own Claude plan, Runs on your own ChatGPT plan, Works inside your own coding agent and You choose which model runs your work

The catch

Official evidence records a gap in published guarantees about your data: no published support for Open source and Run it on your own server.

Pricing

FREEPRO$13/moyrPRO+$39/moyrULTRA$99/moyr

Free–$149/mo · help.notis.ai

Who should pick it

Pick Notis if running long jobs on a machine of its own is what decides it for you. Walk away if you need open source, self-hosting or a named certification.

See what those rows cost80% across 33 vetted rows (all 33)

2. Martin

AI personal assistant reachable by phone, SMS, WhatsApp, email and Slack

A personal assistant you reach by phone call, text, WhatsApp, email or Slack that runs your calendar, inbox, reminders and notes.

Best for: remembering context between conversations

Documented strengths

Where you can use it: WhatsApp, Slack, Email, SMS or a phone number, Voice messages, Web or desktop app and Its own iPhone or Android app
Memory: Remembers you between conversations and Reads your existing files and apps
Runs without you: Runs on a schedule or a trigger and Reminds and nudges you

The catch

Team and plan options: none of it. Nothing published on Free plan does the product's main job and Shared team workspace.

Pricing

Basic$21/moyrPro$30/moyr

Basic $35/mo, $21/mo billed yearly; Pro $49/mo, $30/mo billed yearly; 7 days free · no free tier · trymartin.com

Who should pick it

Martin is the one to take when remembering context between conversations is the deciding factor. Walk away if you need a usable free plan or team seats.

Our take on how it differs from Notis: Martin covers phone, SMS, WhatsApp, email and Slack with live voice calls and outbound calling on your behalf, but not iMessage or Telegram, and bundles one AI model per plan tier with no MCP, API-key or model-choice option, unlike Notis's channel-first, provider-agnostic execution layer.

Visit Martin41% across 33 vetted rows (all 33)

3. Boski

Proactive personal-goals agent

Personal agent that remembers preferences, works between conversations and checks in on progress toward goals.

Best for: scheduled and event-triggered work

Documented strengths

Where you can use it: WhatsApp, iMessage, Email, Voice messages and Web or desktop app
Memory: Remembers you between conversations and Reads your existing files and apps
Runs without you: Runs on a schedule or a trigger and Reminds and nudges you

The catch

The documented gap is messaging channels. We found no published support for Telegram, Slack and SMS or a phone number.

Pricing

Tiers not collected

$5.00/week, charged weekly (Stripe web checkout); 5,000 credits per one-week allowance period; no monthly or annual price published · no free tier · boski.com

Who should pick it

Worth it for scheduled and event-triggered work — that is what Boski is built around. Walk away if you need it in the other apps you message from.

Our take on how it differs from Notis: Proactive personal-goal assistant with eight built-in apps and recurring check-ins.

Visit Boski39% across 31 vetted rows; 2 rows are not checked yet

4. Noah AI

AI executive assistant (scheduling and follow-ups)

An AI executive assistant you text over SMS, WhatsApp, iMessage, Slack or email to schedule meetings, send emails, set reminders and make phone calls for you.

Best for: reaching it from the apps you already message in

Documented strengths

Where you can use it: WhatsApp, iMessage, Slack, Email, SMS or a phone number and Voice messages
Memory: Remembers you between conversations and Reads your existing files and apps
Runs without you: Reminds and nudges you

The catch

Official evidence records a gap in connected tools: no published support for Creates and edits documents, Save your own reusable instructions, Browses the web for you and Posts to social media for you.

Pricing

Early adopter$49/mo

$49/month · no free tier · heynoah.io

Who should pick it

Pick Noah AI if reaching it from the apps you already message in is what decides it for you. Walk away if you need deep integration with the apps you run on.

Our take on how it differs from Notis: Noah is a narrow executive-admin assistant (calendar, email threads you CC it on, reminders, phone calls) on one undisclosed model stack; Notis adds Telegram and a voice channel, lets you choose the model and bring your own subscription or key, and runs a far broader tool and automation layer.

Visit Noah AI34% across 32 vetted rows; 1 row is not checked yet

Where none of them wins

Nothing on this page publishes anything on Bring your own API key, Open source and Run it on your own server. If one of them is the thing you actually need, none of these is your answer.

One more thing, because a table like this can be read for more than it is worth. It counts what each product documents. It says nothing about how reliable any of them is, how fast it replies, how good its writing is, or what it does when an integration breaks on a Friday night. A lower support score can still be the better daily driver. The only way to know is to put a real week of work through both.

A dash means the capability was not documented on the official pages we checked. Not checked means the evidence is still incomplete.

Questions people ask

Which of Martin, Boski or Noah AI scores highest in the evidence checked so far?

Martin, with 41% across 33 vetted rows (all 33). Boski: 39% across 31 vetted rows; 2 rows are not checked yet and Noah AI: 34% across 32 vetted rows; 1 row is not checked yet. Unknown rows stay out of the score, so treat this as an evidence status, not a complete market verdict.

How much do they cost?

Martin: Basic $35/mo, $21/mo billed yearly; Pro $49/mo, $30/mo billed yearly; 7 days free; Boski: $5.00/week, charged weekly (Stripe web checkout); 5,000 credits per one-week allowance period; no monthly or annual price published; Noah AI: $49/month. Different currencies and different units, so there is no honest way to rank these against each other.

Which of them can I try for free?

Martin, Boski and Noah AI do not have one.

Which of them work in iMessage?

Boski and Noah AI do. Martin does not.

What can none of them do?

Not one of them publishes anything on Bring your own API key, Open source and Run it on your own server. A dash means the capability was not documented on the official pages we checked. Not checked means the evidence is still incomplete.

How was this comparison put together?

We read whatever each product publishes — 100 first-party sources across 4 products — and scored all of them on the same 33 rows. Last checked 21 August 2026. A dash means the capability was not documented on the official pages we checked. Not checked means the evidence is still incomplete.

Where does Notis fit in?

At number 1, scored by the same code on the same 33 rows as everything else. Against Martin specifically: Martin covers phone, SMS, WhatsApp, email and Slack with live voice calls and outbound calling on your behalf, but not iMessage or Telegram, and bundles one AI model per plan tier with no MCP, API-key or model-choice option, unlike Notis's channel-first, provider-agnostic execution layer.

How we checked all this

We read every product on this page against whatever it publishes — docs and pricing pages where there are any, the homepage where that is all there is — against the same 33 buyer-focused rows. A support score uses only rows with an explicit vetted state; every unchecked row remains visible and never becomes a no. Notis chose the questions and is scored by the same rule. Count the links below before you weigh a result; some vendors give us three pages and some give us one. Last checked 21 August 2026. No vendor briefings, no affiliate links, no press releases.

Sources

Spotted something out of date? These things move weekly and we would rather be corrected. Tell us at flo@mindtheflo.com and the row gets rechecked.

Stop comparing. Start delegating.

Message Notis from WhatsApp, iMessage, Telegram, Slack, or email and let it carry the work across every tool behind your business.

Start free