Help build local AI with no API token caps.
Founders + one-time chip-ins fund the next model and agent sprint.
Fund it →
v1.11.593 · notarized Mac DMG · community-funded

A private, offline alternative to cloud AI like Claude and ChatGPT.
No caps. No data centers. All on your Mac.

Outlier is a Mac-native beta app for people who are tired of paying $200/month and still hitting usage limits. The goal is simple: cloud-style coding workflows, running locally by default, with no API token caps from Outlier. The app is early. The mission is not.

Outlier is built by one person, in Grand Rapids. No VC, no investors, no API margins to protect. That's why Free is genuinely free, Pro is $20/mo, and lifetime starts at $99. Great product, priced fairly.

🔒 Your prompts and files never leave your Mac. Local models run fully offline — only model downloads and optional web search ever touch the network, and only when you choose. No telemetry on inference. Open-weight models you can audit on HuggingFace.

Not ready to install? Get launch news + major updates by email — no spam, unsubscribe anytime.
Apple Silicon (M1, M2, M3, or M4) required — Intel Macs are not supported. · macOS 12+ · Free Nano + Lite · Pro $20/mo or $149/yr adds everything incl. Plus · lifetime from $99 · local by default · performance varies by hardware
Outlier is actively under development. “Unlimited” means no API token meter from Outlier for local models; speed, context, model fit, and battery life depend on your Mac.
Local usage
$20
Pro / month
$149
Pro / year
$99
Lifetime, from
100%
Local by default
7
Model tiers
0
Cloud token bill

The pitch in one sentence

Cloud coding tools are incredible, but usage caps kill momentum. Outlier is the bet that the same style of coding platform can live on your Mac: agent plans, file edits, tests, memory, vision, research, and long runs without a meter.

Free: Nano + Lite for everyday local AI. $0 forever.
Pro: $20/month or $149/year. Everything, including Plus 397B.
Founding 200: $99 once, lifetime Pro, 200 seats — sells first.
Founders 500: $200 once, lifetime Pro — opens after Founding 200 sells out.

What the money funds

Every dollar goes toward the gap between where Outlier is today and where it needs to be: stronger coding models, better agent loops, faster local inference, cleaner UX, and public build logs showing exactly how we compress, speed up, test, and ship.

Community-funded. Build-in-public. No cloud token tax.

Will it run on your Mac?

16 GB
MacBook Air
Nano + Lite
32 GB
MacBook Pro M-series
+ Quick 26B, Core 27B, Code 27B, Vision 35B
64 GB
Mac Studio / MBP
All tiers, incl. Plus 397B
96+ GB
Mac Studio / MacPro
Plus 397B with more headroom

Apple Silicon only (M1/M2/M3/M4). Intel Macs not supported.

The problem

You can pay $200/month and still run out.

That is the pain Outlier is built around. The best AI coding tools are powerful, but they are rented, rate-limited, and cloud-hosted. Outlier is building the local version: your Mac does the work, your code stays private, and usage does not stop because a server-side meter says so.

Typical cloud AI

Usage caps break flow

Long coding sessions hit limits. The tool gets good right when you need it most, then cuts off. Local inference changes the economics: once the model is downloaded, every token runs on your hardware.

Typical cloud AI

Your code leaves your machine

Private repos, customer data, personal documents, and business plans all move through someone else’s infrastructure. Outlier is local-first: the default path is your Mac, your files, your disk.

Typical cloud AI

You rent the workflow forever

Stop paying, lose access. Outlier’s free tier remains useful forever, Pro is $20/month (or $149/year) for everything including Plus 397B, and lifetime starts at $99 — a one-time purchase that grants Pro for life while funding the product into existence.

Simple pricing

Free to start. $20/mo Pro. Lifetime from $99. Founders fund the mission.

The pricing has one job: keep local AI accessible while raising enough money to build a stronger offline coding-agent platform. Free gets Nano + Lite forever. Pro is $20/month or $149/year and adds everything — all seven model tiers including Plus 397B, memory, projects, web research, MCP, code agent v2, marathon, compare, computer use, vision, voice, and the API server. Lifetime Pro starts at $99 while Founder seats last.

No investors and no API margins to protect — that's why Free is genuinely free, Pro is $20/mo, and lifetime starts at $99.

sells first

Founding 200

$99 once · lifetime Pro
200 seats

The first 200 lifetime seats, at the lowest price local Pro will ever be. Once they fill, the $200 Founders 500 cohort opens. Directly funds the build.

  • Pro forever — every Pro feature
  • Plus 397B included
  • All future model releases
  • Private founding-cohort Discord
  • Founder recognition
  • 200 seat cap — sells first
Claim a Founding 200 seat

Free

$0 forever

For everyone who wants local private AI without a subscription.

  • Nano (4B) + Lite (9B)
  • Local chat + sessions
  • Model picker
  • No account required
  • No token bill
Download beta app

Founders 500

$200 once · lifetime Pro

Opens after Founding 200 fills. Pro forever, 500 lifetime seats. Once they're gone, recurring Pro is the only path.

  • Pro forever — everything + Plus 397B
  • All future model releases
  • Founder recognition + early access
  • Early builds + build logs
  • 500 seat cap — opens after Founding 200
Opens after Founding 200 fills
Why we need the community

The app works. The ambition is bigger.

Outlier doesn't claim to match the best cloud coding agents yet — and honestly, that's the exciting part: today's app is the foundation, the community funds the climb, and every major step ships in public — the compression work, the benchmarks, the failed runs, the speedups, the fixes. It gets better build after build, and you can watch it happen.

Now · v1.11.593

Local Mac AI foundation

Signed, notarized Mac app. Streaming chat, local sessions, seven model tiers, model downloads, memory, project files, web research, vision support, and a Pro gate that turns on the full stack.

Next · agent quality

Cloud-style coding workflow

Plan → diff → approve → patch → test → repair. The goal is not just a chat model that writes code; it is a local coding platform that can work across a repo without burning cloud usage.

Next · model quality

Better coding models

Every founder seat helps pay for the runs, evals, and infrastructure to close the quality gap. Stronger local coding, better reasoning, better long-context behavior, and honest benchmark reporting.

Ongoing · speed

Compression + inference work

The build logs will show how we shrink, route, cache, page, and speed up large open models so they become practical on consumer Macs instead of locked behind H100s.

This is a community bet: enough people want unlimited local AI badly enough to fund the missing pieces together. If that is you, Founders is the cleanest way to help.

7-tier model lineup

Free starts small. Pro adds the other five tiers.

The product is organized around simple expectations: Free is useful immediately with Nano + Lite. Pro ($20/mo or $149/yr) adds the other five tiers — Quick, Core, Code, Vision, and Plus 397B. The lineup below reflects the current shipping tiers.

TierPlanBest forDisk / RAMSpeed / note
Outlier Nano
Qwen3.5-4B · MLX 4-bit
Free Fast iteration, lightweight chat, small Macs 2.37 GB · 6 GB min RAM ~30-70 tok/s (Mac-dependent)
Outlier Lite
Qwen3.5-9B · MLX 4-bit
Free Daily local AI, writing, search, Q&A 5.04 GB · 12 GB min RAM Mac-dependent
Outlier Quick
Gemma-4-26B MoE
Pro Thinking-mode reasoning, not a code substitute 15.61 GB · 16 GB min RAM Mac-dependent
Outlier Core
Qwen3.6-27B · text-only
Pro Best default quality, reasoning, coding 15.13 GB · 24 GB min RAM High-end quality, fully offline
Outlier Code
Core weights + code config
Pro Coding workflow, lower-temp code-tuned setup 15.13 GB · 24 GB min RAM Same verified base as Core
Outlier Plus
Qwen3.5-397B-A17B · V9 paged
Pro Largest local tier · 397B MoE on high-end Macs · only on Pro or Founders 209 GB disk · 64 GB min, 128 GB recommended Trophy tier — capability over speed
Outlier Vision
Qwen3.6-35B-A3B · vision retained
Pro Images, screenshots, OCR, multimodal reasoning 19.0 GB · 24 GB min RAM V9 K=256: 16.31 tok/s @ 34.04 GB

Quick / Core / Code / Vision / Plus 397B are all included with Pro ($20/mo or $149/yr) or lifetime Pro (Founding 200 / Founders 500). Code uses the same weights as Core with code-specialized configuration. Quick is useful for reasoning, not positioned as a coding tier.

What ships now

A local AI workstation, not just a chat box.

Today's build is the foundation: local chat, session history, model picker, model downloads, project context, memory, web research, agent tools with approval, test panels, vision, and a Mac-native app shell. Some edges are rough. That is why the build is funded in public.

Local

Chat + sessions

Streaming token output, persistent local history, rename/delete/pin, Markdown export, demo session on first run, cost display at $0.00, and local storage across restarts.

Agent

Tools with approval

File read/write, shell execution with approval, permission modes, plan review card, repair loop, audit log, path scoping, project map, symbols, dependencies, snapshots, and tests.

Research

Search + citations

Deep research mode, DuckDuckGo with Wikipedia fallback, source filters, export, follow-up, summary cards, trust badges, and inline citations with source excerpts.

Memory

Local memory

Persistent memory in SQLite, short/medium/long-term tiers, provenance tracking, review cards, conflict detection, decay, frequency, and MEMORY.md export.

Vision

Images + screenshots

Image upload and direct image queries through Outlier Vision. Best for OCR, screenshots, diagrams, and multimodal Q&A — not positioned as the coding model.

Mac

Signed + notarized

macOS arm64 DMG, Apple notarization accepted, Gatekeeper accepted, GitHub Releases distribution, and a Tauri updater pointed at the latest manifest.

The comparison that matters

Outlier vs. the monthly bill.

The cloud tools proved the workflow: coding agents, long-context research, file-aware assistants, and always-on help. The problem is the meter. Outlier is building the local version: Mac-native, private by default, and not capped by an Outlier API token allowance.

Cloud AI tools Cloud coding agents Outlier Free Outlier Pro
Monthly cost $20+ Often much higher $0 $20/mo (or $149/yr · lifetime from $99)
Usage model Server-side limits Usage windows / caps No Outlier token meter No Outlier token meter
Where inference runs Provider cloud Provider cloud Your Mac Your Mac
Privacy default Remote request Remote repo/context Local by default Local by default
Offline use No No Yes, once models download Yes, once models download
Current maturity Very mature Very mature Useful beta Ambitious beta
Goal Hosted assistant Hosted coding workflow Local baseline Cloud-style coding workflows, local

Important framing: Outlier is not claiming parity with the best cloud coding agents today. The beta is the foundation; Pro and Founders revenue funds the climb toward that experience locally.

Why local matters

Local AI can be lighter on the planet.

Cloud AI bills you twice — your wallet and the planet. Cloud inference needs datacenters, networking, cooling, and always-growing GPU clusters. Local inference uses the Apple Silicon chip you already own — no datacenter round-trip, and no per-token meter. The cloud bill is real and rising: 2025 cost reports (e.g. CloudZero's State of AI Costs) describe enterprise AI spend climbing fast, with agentic, multi-step workflows burning far more tokens than teams budget for. Outlier's price is fixed and your local usage isn't metered.

80–90%

of AI compute is inference — the token calls made every time someone uses a cloud model, now drawing more grid electricity than training did (MIT Tech Review, 2025). Outlier runs them on the Mac you already own.

0 cloud tokens

Outlier local models do not create an Outlier cloud inference bill or cloud token meter.

Mac

Apple Silicon unified memory is efficient for local inference compared with shipping every prompt to a server.

This is not a claim that every local query is automatically cleaner in every situation. Hardware, model size, electricity source, and usage pattern all matter. The point is directionally important: if a huge share of everyday AI inference can move from datacenters to efficient devices people already own, the load on cloud infrastructure can drop.

That is why compression, routing, quantization, paging, and model fit are not just engineering details. They are part of the product philosophy. A useful local model is not just cheaper for the user — it can also reduce unnecessary cloud dependence for everyday tasks.

The best environmental feature is not a green badge. It is a model that is good enough, small enough, and fast enough that people actually choose to run it locally.

Practical framing: local-first when possible, cloud only when needed. Outlier’s default path is the Mac.

Proof, not vibes

Numbers with a paper trail.

The old website had a strong provenance section. It belongs here. Outlier should be ambitious, but the numbers still need to be honest: source file, command, sample size, standard error, date measured.

📁 Source files

Benchmarks should trace back to eval artifacts or sprint logs. If a number cannot point to a file, it should not become marketing copy.

⌨️ Exact commands

The harness, version, dtype, device, shot count, and sample size matter. Smoke tests stay smoke tests.

📊 Standard error

Small benchmark lifts are not magic. Outlier’s public story should separate strong absolute scores from statistically significant improvements.

🚫 No fake parity

The goal is cloud-style coding workflows running locally. The current beta is not yet equal to the best cloud coding agents, and the site should say that clearly.

🧪 Public build process

The community funds model runs, evals, compression, speed work, agent loops, and UX hardening. The process is part of the product.

⚡ One-month pace

Outlier went from idea to shipped beta in about a month. More runway means more focused cycles on model quality, agents, and polish.

Current site-safe headline numbers: v1.11.593 shipped as a notarized Mac DMG; Free includes Nano + Lite; Pro ($20/mo or $149/yr) adds everything including Plus 397B; lifetime Pro from $99 (Founding 200) or $200 (Founders 500); local generation has no Outlier API token meter.

Research + build track

Where the next money goes.

Founders and chip-ins are not abstract support. They fund specific work: better coding models, better local speed, more reliable agents, better tests, and a public process that shows the wins and failures.

Shipped · v1.11.593

Mac-native local foundation

Notarized DMG, seven model tiers, local chat, sessions, downloads, memory, project context, research, vision, and Pro upgrade.

● Live beta
Next · agent workflow

Plan → diff → test → repair

The product goal is a local coding-agent loop that can work across real projects without burning paid cloud usage.

● Funded by Pro + Founders
Next · model quality

Stronger local coding tiers

Better coding performance means disciplined evals, better prompts/configs, distillation experiments, and no inflated benchmark claims.

● Community-funded
Ongoing · speed

Compression + inference

Make bigger open models practical on consumer Macs through paging, caching, quantization, routing, and product-level hardware honesty.

Always improving

The pitch is not “trust us.” The pitch is: help fund the work, use the beta, report what breaks, and watch the process of making local AI better in public.

Fund the build

Help us build the platform we all want to use.

This is the call to action: buy Founders, subscribe to Pro, or chip in whatever you can. The money goes into making Outlier better — coding-agent quality, local model quality, speed, compression, and the public process behind all of it.

Outlier went from idea to a shipped, notarized Mac app in about a month. That included seven model tiers, local inference, agents, memory, vision, payments, packaging, and the first public build. Imagine what this can become with more runway and two more months of focused building. We will take anything you can give us. Thank you.

Best way to help · lifetime Pro

Founders

from $99 · one time

Lifetime Pro — everything Outlier does, forever. Founding 200 is $99 once (200 seats, sells first); Founders 500 is $200 once and opens after Founding 200 sells out. Early builds. Founder recognition. A direct vote for local AI with no API token caps. Once the lifetime seats are gone, recurring Pro is the only path.

Become a Founder →
Pro
$20/mo · $149/yr
Everything Outlier does, including Plus 397B. Yearly saves $91 vs monthly.
Subscribe →
Chip in
Any amount
Chip in anything you can give. One time. No subscription. Thank you — genuinely.
Chip in →
Custom support
Any amount
Want to sponsor a benchmark run, feature, or larger amount? Email Matt directly.
Email Matt →
Community

We need everybody’s help.

The community is not a side quest. It is how this gets built: users testing real workflows, reporting what breaks, funding the next sprint, and holding the benchmarks honest. The more people using it, the faster it gets better.

FAQ

Clear answers.

Is Outlier already as good as the best cloud coding agents?
No. That is the goal, not the current claim. Outlier is a real local Mac app with shipped models and agent features, but the app and models still need work to reach the quality of the best cloud coding agents. Founders and Pro revenue fund that gap.
What does “unlimited tokens” mean?
Outlier does not meter local model usage or charge per token. Once a model is downloaded, generation runs on your Mac. Your practical limit is hardware, power, storage, model size, and time — not a server-side usage cap from Outlier.
What is free?
Nano and Lite. They are the lightweight local tiers for everyday use. Free is meant to be useful, not a fake trial.
What does Pro include?
Pro is $20/month or $149/year (yearly saves $91 vs monthly). Pro includes everything — all seven model tiers including Plus 397B (Outlier's largest local model, running paged-MoE on Apple Silicon), plus memory, projects, web research, MCP, code agent v2, marathon, compare, computer use, vision, voice, API server, and long context. There is no middle tier — it's Free or Pro.
Is there a lifetime option?
Yes. Founding 200 is $99 once for lifetime Pro — 200 seats, sells first. Founders 500 is $200 once for lifetime Pro and opens after Founding 200 sells out. Both grant everything Pro does, forever, including Plus 397B and all future model releases. Once the lifetime seats are gone, recurring Pro is the only path.
Does my data stay private?
Local model inference runs on your Mac. Chat history and memory are stored locally. Web search and external APIs are optional paths, not the default. If you turn on web search or bring an external API key, that specific request leaves your device by design.
What Mac do I need?
Apple Silicon Mac. Nano and Lite fit smaller machines; Core, Code, and Vision are best on 24 GB+ RAM; Plus needs significant disk space and is best for higher-end Macs.
Can I just donate?
Yes. Use the chip-in link for a one-time contribution of whatever you can give, or email Matt for a custom sponsorship, benchmark run, or feature sponsorship.
What doesn't work yet? (honest list)
Deep Research v3 abort: clicking Stop on a long research run releases the UI immediately, but the backend's current generation finishes draining in the background (~5–15s). Fixed when mlx_lm upstream adds a stop-callback primitive we don't own. Workaround: /restart if it wedges.

Plus first-token latency: 60–90s on 64 GB Macs for cold prefill on the 397B MoE. Use Code 27B for everyday chat; Plus shines on deep thinking where the wait is worth it.

Intel Macs not supported: Apple Silicon only. MLX is the reason Outlier is fast — there's no Intel path.

Some surfaces still rough: Companion overlay is a v0; Knowledge Stacks uses a hash-based retriever (real embedder lands v1.12); Vision multi-turn just got fixed in v169 so prior chats may need a fresh start.
Questions before buying? Who do I email?
matt@outlier.host — every email reaches Matt directly. Usually answered within a few hours during working hours (US Eastern).
Build the offline future

Unlimited local usage. Private by default.

The cloud tools proved what the workflow should feel like. Now we build the version that runs locally, belongs to the user, and has no API token cap from Outlier mid-sprint.

Requires a Mac with Apple Silicon (M1, M2, M3, or M4) — Intel Macs are not supported. macOS 12+.