Skip to content

Llama 3.3 70B on 2× RTX 5090: VRAM, speed and fit

Llama 3.3 70B at 6-bit runs on the 2× RTX 5090 with 16K of context at about 21 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on 2× RTX 5090 · 6-bit (Q6_K) · Ollama / LM Studio · Chat

GPU memory
60.5 GB
of 64.0 GB
Context
16K
auto
Speed
21 tok/s
Fast
First reply
0.8 s
2K prompt

Where the memory goes

Model
53.9 GB
Context
5.0 GB
Overhead
1.6 GB
OS
0.0 GB
Free
3.5 GB
Memory
60.5 GB of 64.0 GB
Context
16K tokens · auto
Speed
21 tok/s · fast
First reply
0.8 s for a 2K prompt
Running cost
$17/mo at home · $117/mo on a cloud 4× RTX 2000 Ada

Validation: Validated

What Llama 3.3 70B on 2× RTX 5090 was checked against
CheckTargetResultStatus
Fits in GPU memory64.0 GB usable60.5 GB neededPass
Headroom5% free3.5 GB free (5%)Pass
Context for chat16K or more16KPass
Writing speed15 tok/s or more21 tok/sPass
First replyunder 3 s, 2K prompt0.8 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
Llama 3.3 70B
Parameters
70.5B
Precision
6-bit (Q6_K)
Context
16K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
2× RTX 5090
GPU can use
64.0 GB of 64.0 GB

Worth trying

  • 8-bit cache: Room for 28K context

Running cost

At home
$17/mo
770 W at the wall
Renting instead
$117/mo
4× RTX 2000 Ada at $0.96/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 2× RTX 5090 run?

Other models that run on 2× RTX 5090

Questions

Can Llama 3.3 70B run on 2× RTX 5090?

Llama 3.3 70B at 6-bit runs on the 2× RTX 5090 with 16K of context at about 21 tokens a second.

Report VR-09OFR2H, calculated Nov 22, 2026 from published specs, not benchmarked.