Skip to content

Granite 3.3 8B Instruct on RTX 3090: VRAM, speed and fit

Granite 3.3 8B Instruct at 8-bit runs on the RTX 3090 with 88K of context at about 69 tokens a second.

Text generation · Chat · Ollama / LM Studio

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

GPU memory
22.5 GB
of 24.0 GB
Context
88K
auto
Speed
69 tok/s
27 when full
First reply
0.5 s
2K prompt

Where the memory goes

Model
8.1 GB
Context
13.8 GB
Overhead
685 MB
OS
0.0 GB
Free
1.5 GB
Memory
22.5 GB of 24.0 GB
Context
88K tokens · auto
Speed
69 tok/s · 27 when full
First reply
0.5 s for a 2K prompt
Running cost
$6.35/mo at home · $33/mo on a cloud RTX A5000

Validation: Validated

What Granite 3.3 8B Instruct on RTX 3090 was checked against
CheckTargetResultStatus
Fits in GPU memory24.0 GB usable22.5 GB neededPass
Headroom5% free1.5 GB free (6%)Pass
Context for chat16K or more88KPass
Writing speed15 tok/s or more69 tok/sPass
First replyunder 3 s, 2K prompt0.5 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
Granite 3.3 8B Instruct
Parameters
8.2B
Precision
8-bit (Q8_0)
Context
88K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
RTX 3090
GPU can use
24.0 GB of 24.0 GB

Running cost

At home
$6.35/mo
290 W at the wall
Renting instead
$33/mo
RTX A5000 at $0.27/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 ibm-granite/granite-3.3-8b-instruct-GGUF
vLLM
vllm serve ibm-granite/granite-3.3-8b-instruct
MLX (Mac)
mlx_lm.chat --model mlx-community/granite-3.3-8b-instruct-4bit

Granite 3.3 8B Instruct on similar machines

All machines: Granite 3.3 8B Instruct VRAM requirements. Everything that runs here: what can the RTX 3090 run?

Other models that run on RTX 3090

Questions

Can Granite 3.3 8B Instruct run on RTX 3090?

Granite 3.3 8B Instruct at 8-bit runs on the RTX 3090 with 88K of context at about 69 tokens a second.

Report VR-10SFXLO, calculated Nov 22, 2026 from published specs, not benchmarked.