Skip to content

Qwen2.5 72B Instruct on 4× RTX 3090: VRAM, speed and fit

Qwen2.5 72B Instruct at 8-bit runs on the 4× RTX 3090 with 32K of context at about 8.5 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on 4× RTX 3090 · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
85.1 GB
of 96.0 GB
Context
32K
auto
Speed
8.5 tok/s
Reading speed
First reply
1.3 s
2K prompt

Where the memory goes

Model
71.9 GB
Context
10.0 GB
Overhead
3.2 GB
OS
0.0 GB
Free
10.9 GB
Memory
85.1 GB of 96.0 GB
Context
32K tokens · auto
Speed
8.5 tok/s · reading speed
First reply
1.3 s for a 2K prompt
Running cost
$20/mo at home · $131/mo on a cloud 4× RTX A5000

Validation: Validated with conditions

What Qwen2.5 72B Instruct on 4× RTX 3090 was checked against
CheckTargetResultStatus
Fits in GPU memory96.0 GB usable85.1 GB neededPass
Headroom5% free10.9 GB free (11%)Pass
Context for chat16K or more32KPass
Writing speed15 tok/s or more8.5 tok/sCheck
First replyunder 3 s, 2K prompt1.3 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
Qwen2.5 72B Instruct
Parameters
72.7B
Precision
8-bit (Q8_0)
Context
32K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
4× RTX 3090
GPU can use
96.0 GB of 96.0 GB

Running cost

At home
$20/mo
920 W at the wall
Renting instead
$131/mo
4× RTX A5000 at $1.08/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 bartowski/Qwen2.5-72B-Instruct-GGUF
vLLM
vllm serve Qwen/Qwen2.5-72B-Instruct
MLX (Mac)
mlx_lm.chat --model mlx-community/Qwen2.5-72B-Instruct-4bit

Qwen2.5 72B Instruct on similar machines

All machines: Qwen2.5 72B Instruct VRAM requirements. Everything that runs here: what can the 4× RTX 3090 run?

Other models that run on 4× RTX 3090

Questions

Can Qwen2.5 72B Instruct run on 4× RTX 3090?

Qwen2.5 72B Instruct at 8-bit runs on the 4× RTX 3090 with 32K of context at about 8.5 tokens a second.

Report VR-1RS6BPF, calculated Nov 22, 2026 from published specs, not benchmarked.