Skip to content

Llama 3.3 70B on M3 Ultra 96 GB: VRAM, speed and fit

Llama 3.3 70B at 6-bit runs on the M3 Ultra 96 GB with 44K of context at about 8.8 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M3 Ultra 96 GB · 6-bit (Q6_K) · Ollama / LM Studio · Chat

GPU memory
92.4 GB
of 96.0 GB
Context
44K
auto
Speed
8.8 tok/s
7.1 when full
First reply
13 s
2K prompt

Where the memory goes

Model
53.9 GB
Context
13.8 GB
Overhead
763 MB
OS
24.0 GB
Free
3.6 GB
Memory
92.4 GB of 96.0 GB
Context
44K tokens · auto
Speed
8.8 tok/s · 7.1 when full
First reply
13 s for a 2K prompt
Running cost
$3.55/mo at home · $122/mo on a cloud 4× RTX A4500

Validation: Not validated

What Llama 3.3 70B on M3 Ultra 96 GB was checked against
CheckTargetResultStatus
Fits in GPU memory72.0 GB usable68.4 GB neededPass
Headroom5% free3.6 GB free (5%)Pass
Context for chat16K or more44KPass
Writing speed15 tok/s or more8.8 tok/sCheck
First replyunder 3 s, 2K prompt13 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
Llama 3.3 70B
Parameters
70.5B
Precision
6-bit (Q6_K)
Context
44K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
M3 Ultra 96 GB
GPU can use
72.0 GB of 96.0 GB

Worth trying

  • 8-bit cache: Room for 80K context

Running cost

At home
$3.55/mo
162 W at the wall
Renting instead
$122/mo
4× RTX A4500 at $1.00/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

Ollama
ollama run llama3.3:70b
llama.cpp
llama-server -hf unsloth/Llama-3.3-70B-Instruct-GGUF
vLLM
vllm serve meta-llama/Llama-3.3-70B-Instruct
MLX (Mac)
mlx_lm.chat --model mlx-community/Llama-3.3-70B-Instruct-4bit

Llama 3.3 70B on similar machines

All machines: Llama 3.3 70B VRAM requirements. Everything that runs here: what can the M3 Ultra 96 GB run?

Other models that run on M3 Ultra 96 GB

Questions

Can Llama 3.3 70B run on M3 Ultra 96 GB?

Llama 3.3 70B at 6-bit runs on the M3 Ultra 96 GB with 44K of context at about 8.8 tokens a second.

Report VR-1LLO7YG, calculated Nov 22, 2026 from published specs, not benchmarked.