Local AI
Cloud AI / Local AI
What it means to keep AI on your Mac.
There is cloud AI, which runs on servers across the internet, and local AI, which runs right inside your Mac. Each has its own strengths and weaknesses. Here is how they differ, and how R.E.V.I.S. works to overcome the weaknesses of local AI.
CLOUD AI
- You
- Internet
- Service's servers
Your input is processed on the server.
LOCAL AI — R.E.V.I.S.
- You
- Your Mac
Both thinking and remembering happen inside your Mac.
The Difference
The difference is where it thinks.
It may all be called AI, but where it runs changes where your data goes, how you pay, and what you need.
| Cloud AI | Local AI (R.E.V.I.S.) | |
|---|---|---|
| Where it thinks | The service's servers | Your Mac |
| Data you enter | Travels over the internet to the service. | Never leaves your Mac. |
| Pricing | Monthly subscription or pay-as-you-go. | One-time purchase. Using it more never costs more. |
| Model size | Even the largest models are available. | Up to the size that fits in your Mac's memory. |
| What you need | A device connected to the internet. | A Mac with plenty of memory. |
| When things change | Models and usage limits can change at the service's discretion. | The model you use is the one you chose. |
Strengths
What becomes possible when it's on your Mac.
Your content never leaves
Confidential documents and personal questions alike are handled inside your Mac. They are never used as training data.
How we approach safetyUsing it more doesn't cost more
A one-time purchase. Read long documents as many times as you like or ask as many questions as you want — the cost never goes up.
Stays the partner you know
The model you use won't be swapped out behind the scenes, and your usage won't be throttled.
Grows right on your Mac
Learned preferences, past conversations, and things it has researched all build up on your Mac.
Weaknesses — and Our Answers
Honest about the weaknesses. Built to overcome them.
Local AI has clear weaknesses. R.E.V.I.S. tackles each one head-on, starting from the engine itself.
-
01
Models tend to be smaller
Because the model has to fit in your Mac's memory, you use smaller models than the largest cloud AI.
Our custom Q4T and Q3T quantization shrinks models while keeping accuracy close to the original, letting a 35B-class model run on a Mac with 24GB of memory. J.I.T.R.X. tailors prompts for each model to bring out what it does best.
More about quantization -
02
Speed depends on your Mac
Instead of dedicated machines in a data center, calculations run on the Mac in front of you.
Q.E.L.V.A.R.I., an inference engine written from scratch exclusively for the Mac, turns the power of Apple silicon directly into speed. With MTP, where a small draft model writes ahead, answers move forward in fewer steps.
See the speed in numbers -
03
A single machine has limits
How much work you can run at once is bounded by the memory and compute power of a single Mac.
When multiple Macs are on the same network, F.A.L.Q. automatically shares work out to idle Macs.PRO
More about F.A.L.Q. -
04
Knowledge stops at training time
A model doesn't know about events that happened after it was created.
When needed, it searches the Web before answering and includes links to the pages it read. The contents of your conversations are never sent.
-
05
Longer conversations get heavier
As a conversation grows longer, there is more to re-read, and answers get slower.
Context it has finished reading is kept encrypted on your Mac, so when you continue, it only reads what's new. Conversations are kept separate by topic, so unrelated talk never bloats the context.
-
06
Setup takes effort
Most local AI tools require you to set up runtime environments and models yourself.
No need to set up Python or any other runtime environment. Once you install the app, just pick a model under "Download" in Settings. Before loading, it also checks whether the model will fit in your Mac's memory.
Honestly
Where it still can't match the cloud.
It won't have the exact same intelligence as the largest cloud AI. What R.E.V.I.S. aims for is an AI you can trust with your everyday work without ever sending your data outside. Through craft in both the engine and the models, we are narrowing that gap step by step.
Keep AI on your Mac.
Overcome the weaknesses through craft.
SEVEN CORES / ONE CONSCIOUSNESS