Skip to content

VLX Seek 1.5 10B on Mac mini M4 32 GB: VRAM, speed and fit

VLX Seek 1.5 10B at 8-bit runs on the M4 32 GB with 256K of context at about 10.0 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M4 32 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
29.5 GB
of 32.0 GB
Context
256K
auto
Speed
10.0 tok/s
5.3 when full
First reply
11 s
2K prompt

Where the memory goes

Model
9.9 GB
Context
8.0 GB
Overhead
869 MB
OS
10.7 GB
Free
2.5 GB
Memory
29.5 GB of 32.0 GB
Context
256K tokens · auto
Speed
10.0 tok/s · 5.3 when full
First reply
11 s for a 2K prompt
Running cost
$0.85/mo at home · $30/mo on a cloud RTX A4500

Validation: Not validated

What VLX Seek 1.5 10B on Mac mini M4 32 GB was checked against
CheckTargetResultStatus
Fits in GPU memory21.3 GB usable18.8 GB neededPass
Headroom5% free2.5 GB free (12%)Pass
Context for chat16K or more256KPass
Writing speed15 tok/s or more10.0 tok/sCheck
First replyunder 3 s, 2K prompt11 sFails

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
VLX Seek 1.5 10B
Parameters
10B
Precision
8-bit (Q8_0)
Context
256K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
M4 32 GB
GPU can use
21.3 GB of 32.0 GB

Running cost

At home
$0.85/mo
39 W at the wall
Renting instead
$30/mo
RTX A4500 at $0.25/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 Asher-1/VLX-Seek-1.5-10B-GGUF
vLLM
vllm serve omlab/VLX-Seek-1.5-10B

VLX Seek 1.5 10B on similar machines

All machines: VLX Seek 1.5 10B VRAM requirements. Everything that runs here: what can the Mac mini M4 32 GB run?

Other models that run on Mac mini M4 32 GB

Questions

Can VLX Seek 1.5 10B run on Mac mini M4 32 GB?

VLX Seek 1.5 10B at 8-bit runs on the M4 32 GB with 256K of context at about 10.0 tokens a second.

Report VR-0UEZDQF, calculated Nov 22, 2026 from published specs, not benchmarked.