Skip to content

Llama 4 Scout 17B 16E Instruct on M5 Ultra 512 GB: VRAM, speed and fit

Llama 4 Scout 17B 16E Instruct at 8-bit runs on the M5 Ultra 512 GB with 1.3M of context at about 32 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M5 Ultra 512 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
488 GB
of 512 GB
Context
1.3M
auto
Speed
32 tok/s
2.6 when full
First reply
3.3 s
2K prompt

Where the memory goes

Model
108 GB
Context
252 GB
Overhead
811 MB
OS
128 GB
Free
23.7 GB
Memory
488 GB of 512 GB
Context
1.3M tokens · auto
Speed
32 tok/s · 2.6 when full
First reply
3.3 s for a 2K prompt
Running cost
$5.06/mo at home · $574/mo on a cloud 8× A40

Validation: Validated with conditions

What Llama 4 Scout 17B 16E Instruct on M5 Ultra 512 GB was checked against
CheckTargetResultStatus
Fits in GPU memory384 GB usable360 GB neededPass
Headroom5% free23.7 GB free (6%)Pass
Context for chat16K or more1.3MPass
Writing speed15 tok/s or more32 tok/sPass
First replyunder 3 s, 2K prompt3.3 sCheck

Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.

The setup

Model
Llama 4 Scout 17B 16E Instruct
Parameters
109B, 18B active
Precision
8-bit (Q8_0)
Context
1.3M, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
M5 Ultra 512 GB
GPU can use
384 GB of 512 GB

Running cost

At home
$5.06/mo
231 W at the wall
Renting instead
$574/mo
8× A40 at $4.72/h

4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 22, 2026.

How to run it

llama.cpp
llama-server -hf unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF
vLLM
vllm serve meta-llama/Llama-4-Scout-17B-16E-Instruct

Llama 4 Scout 17B 16E Instruct on similar machines

All machines: Llama 4 Scout 17B 16E Instruct VRAM requirements. Everything that runs here: what can the M5 Ultra 512 GB run?

Other models that run on M5 Ultra 512 GB

Questions

Can Llama 4 Scout 17B 16E Instruct run on M5 Ultra 512 GB?

Llama 4 Scout 17B 16E Instruct at 8-bit runs on the M5 Ultra 512 GB with 1.3M of context at about 32 tokens a second.

Report VR-18NM8T9, calculated Nov 22, 2026 from published specs, not benchmarked.