Skip to content

Ornith 1.5 122B on M5 Ultra 512 GB: VRAM, speed and fit

Ornith 1.5 122B at 8-bit runs on the M5 Ultra 512 GB with 256K of context at about 41 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
259 GB
of 512 GB
Context
256K
auto
Speed
41 tok/s
31 when full
First reply
2.3 s
2K prompt

Where the memory goes

Model
124 GB
Context
6.1 GB
Overhead
837 MB
OS
128 GB
Free
253 GB
Memory
259 GB of 512 GB
Context
256K tokens · auto
Speed
41 tok/s · 31 when full
First reply
2.3 s for a 2K prompt
Running cost
$5.06/mo at home · $287/mo on a cloud 4× A40

Validation: Validated

What Ornith 1.5 122B on M5 Ultra 512 GB was checked against
CheckTargetResultStatus
Fits in GPU memory384 GB usable131 GB neededPass
Headroom5% free253 GB free (66%)Pass
Context for chat16K or more256KPass
Writing speed15 tok/s or more41 tok/sPass
First replyunder 3 s, 2K prompt2.3 sPass

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
Ornith 1.5 122B
Parameters
125B, 12.7B active
Precision
8-bit (Q8_0)
Context
256K, 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
$287/mo
4× A40 at $2.36/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/Ornith-1.5-122B-GGUF
vLLM
vllm serve ornith-ai/Ornith-1.5-122B

Ornith 1.5 122B on similar machines

All machines: Ornith 1.5 122B 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 Ornith 1.5 122B run on M5 Ultra 512 GB?

Ornith 1.5 122B at 8-bit runs on the M5 Ultra 512 GB with 256K of context at about 41 tokens a second.

Report VR-1U1COZ4, calculated Nov 22, 2026 from published specs, not benchmarked.