Ollama hosting on a dedicated Mac mini or Mac Studio
Macyou rents dedicated Apple Silicon Macs for Ollama, with administrator access, from $99/mo for an M4 Mac mini up to an M3 Ultra Mac Studio with 256 GB of unified memory. Each machine is used by one customer, is ready in about 5 minutes, and runs your models around the clock at a fixed monthly price with no per token fees.
Which Mac runs which model
On Apple Silicon the model has to fit in unified memory, and generation speed follows memory bandwidth. Measured figures come from our fleet benchmarks; estimates use the same formula, which stays within about 5% of every machine we have measured.
| Machine | Fits (Q4) | Typical speed | Monthly |
|---|---|---|---|
| M4 Mac mini, 16 GB | 3B to 14B models | Llama 3.1 8B: 21.2 tok/s (measured) | from $99/mo |
| M4 Mac mini, 32 GB | 14B comfortably, 32B at the edge | Qwen 2.5 14B: 11.7 tok/s (measured) | from $171/mo |
| M4 Pro Mac mini, 48 GB | 32B models with long context | 32B: about 12.3 tok/s (estimate) | from $257/mo |
| M4 Pro Mac mini, 64 GB | 70B at Q4 | 70B: about 5.7 tok/s (estimate) | from $286/mo |
| M4 Max Mac Studio, 128 GB | 70B at Q8, 123B at Q4 | 70B: about 11.3 tok/s (estimate) | from $471/mo |
| M3 Ultra Mac Studio, 256 GB | 200B class at Q4, 70B at FP16 | 70B: about 16.6 tok/s (estimate) | from $828/mo |
For any other model, quantization or context length, the Mac LLM calculator works out memory and speed. Yearly billing is cheaper: see pricing.
What comes on the machine
- Ollama with the model of your choice: Llama, Qwen, Mistral, DeepSeek and others
- Ollama’s OpenAI compatible endpoint on your own Mac, reachable through your SSH connection
- Anything else you install as the administrator: MLX, Jupyter, VS Code Server, agent frameworks such as CrewAI and LangGraph
- SSH, and the full screen through Screen Sharing or any VNC viewer
- A physical Mac nobody else uses, with a hardware encrypted SSD
Ollama, LM Studio, MLX or llama.cpp
The same Mac runs any of the four; the model and the memory stay the same, only the way you call it changes.
| Runtime | Good for | On a Macyou machine |
|---|---|---|
| Ollama | An API for apps and agents, one command per model | One command to install; serves an OpenAI compatible endpoint |
| LM Studio | A desktop chat window and a model browser | Runs on the Mac’s screen through Screen Sharing; its headless server also answers OpenAI style requests. Setup guide |
| MLX | Apple's own machine learning framework for Apple Silicon | Install with pip; mlx_lm.server serves a model over HTTP |
| llama.cpp | The C++ engine under many tools, with full control over quantization | Install over SSH with Homebrew; llama-server exposes an OpenAI compatible API |
Calling your model
Open an SSH tunnel to Ollama’s port, then point any OpenAI client at it:
# ssh -fN -L 11434:localhost:11434 tenant@your-mac
from openai import OpenAI
client = OpenAI(
api_key="ollama", # Ollama ignores the key
base_url="http://localhost:11434/v1",
)
reply = client.chat.completions.create(
model="llama3.3:70b",
messages=[{"role": "user", "content": "Hello!"}],
)
print(reply.choices[0].message.content)Ollama documents this endpoint as its OpenAI compatibility API.
Compared with the usual Ollama picks
The largest M4 Pro each provider sells, against the same machine at Macyou. On each of them, as on Macyou, you install Ollama yourself.
| Provider | Machine | Their price | Same machine at Macyou |
|---|---|---|---|
| MacStadium | M4 Pro, 48 GB, 1 TB | $349/mo | $286/mo, Ollama and API included |
| AWS EC2 Mac | M4 Pro, 48 GB, 1 TB | $1,438/mo | $286/mo, Ollama and API included |
| Scaleway | M4 Pro, 64 GB, 2 TB | $369/mo | $371/mo, Ollama and API included |
| My Remote Mac | M4 Pro, 48 GB, 1 TB | $299/mo | $286/mo, Ollama and API included |
| JUUZ Cloud | M4 Pro, 48 GB, 512 GB | $351/mo | $257/mo, Ollama and API included |
| RoundFleet | M4 Pro, 24 GB, 512 GB | $199/mo | $200/mo, Ollama and API included |
| MacWeb | M4 Pro, 64 GB, 1 TB | $399/mo | $314/mo, Ollama and API included |
| Mac Mini Vault | M4 Pro, 64 GB, 1 TB | $274.99/mo | $314/mo, Ollama and API included |
The M4 Mac mini at other rental sites
An M4 Mac mini with 16 GB is enough for 7B to 14B models. At Macyou it is $99 a month billed monthly or $80 a month billed annually, with Ollama and the API already set up.
| Provider | Billed monthly | Longer terms | Checked |
|---|---|---|---|
| RentAMac | $119 | $639 for 6 months ($106.50 a month) | 2026-10-16 |
| Mikegyver | $99 | 100 Mbps network; $114 with 500 Mbps | 2026-10-19 |
| RunMini | $109 | $299 a quarter ($99.67 a month) | 2026-10-16 |
Every provider side by side, with prices and terms: Mac cloud hosting compared.
Frequently asked questions
Can I host Ollama on a Mac in the cloud?
Yes. A Macyou machine is a dedicated Mac mini or Mac Studio on which you are the administrator, so Ollama installs with one command and serves its OpenAI compatible API on the Mac. It is ready about 5 minutes after your payment is confirmed, and you reach it over SSH or through Screen Sharing.
Which Mac do I need for Llama 70B?
At Q4 a 70B model needs about 42 GB for weights plus room for context, so 64 GB is the entry point. An M4 Pro with 64 GB generates about 5.7 tokens per second; an M4 Max Mac Studio roughly doubles that, and an M3 Ultra with 256 GB also runs 70B at full precision.
How fast is Ollama on an M4 Mac mini?
On our fleet an M4 Mac mini with 16 GB generates 21.2 tokens per second on Llama 3.1 8B at Q4_K_M, measured with Ollama 0.31.2. Speed follows memory bandwidth, so the Pro, Max and Ultra chips are proportionally faster.
Is the API compatible with the OpenAI SDK?
Yes. Ollama’s endpoint follows the OpenAI request format (/v1/chat/completions, /v1/models), so existing code works after you change the base URL. The model itself runs on your machine, and you reach the endpoint on the Mac or through your SSH connection; no request goes to OpenAI.
Can I use MLX, llama.cpp or LM Studio instead of Ollama?
Yes. You are the administrator, so you install MLX, Jupyter, llama.cpp, LM Studio’s headless server or any other runtime yourself.
Why a Mac instead of a GPU cloud?
Unified memory lets one machine hold models that would need several consumer GPUs, at a fixed monthly price with no per token or per hour billing. For a model that has to answer around the clock, that usually costs less than renting GPUs by the hour.
Prefer a desktop app to an API? LM Studio on a cloud Mac runs on the same machines. Moving from MacStadium? See MacStadium alternatives.