Skip to content

Laguna S 2.1 on DGX Station GB300: VRAM, speed and fit

Laguna S 2.1 at 8-bit runs on the DGX Station GB300 with 1M of context at about 213 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on DGX Station GB300 · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
165 GB
of 252 GB
Context
1M
auto
Speed
213 tok/s
66 when full
First reply
0.1 s
2K prompt

Where the memory goes

Model
116 GB
Context
48.1 GB
Overhead
753 MB
OS
0.0 GB
Free
86.9 GB
In RAM
0.0 GB
Memory
165 GB of 252 GB
Context
1M tokens · auto
Speed
213 tok/s · 66 when full
First reply
0.1 s for a 2K prompt
Running cost
$23/mo at home · $287/mo on a cloud 4× A40

Validation: Validated

What Laguna S 2.1 on DGX Station GB300 was checked against
CheckTargetResultStatus
Fits in GPU memory252 GB usable165 GB neededPass
Headroom5% free86.9 GB free (34%)Pass
Context for chat16K or more1MPass
Writing speed15 tok/s or more213 tok/sPass
First replyunder 3 s, 2K prompt0.1 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
Laguna S 2.1
Parameters
118B, 8.4B active
Precision
8-bit (Q8_0)
Context
1M, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
DGX Station GB300
GPU can use
252 GB of 252 GB
Spill space
480 GB LPDDR5X on the Grace CPU

Running cost

At home
$23/mo
1040 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 21, 2026.

How to run it

llama.cpp
llama-server -hf poolside/Laguna-S-2.1-GGUF
vLLM
vllm serve poolside/Laguna-S-2.1
MLX (Mac)
mlx_lm.chat --model mlx-community/Laguna-S-2.1-oQ4e

Laguna S 2.1 on similar machines

All machines: Laguna S 2.1 VRAM requirements. Everything that runs here: what can the DGX Station GB300 run?

Other models that run on DGX Station GB300

Questions

Can Laguna S 2.1 run on DGX Station GB300?

Laguna S 2.1 at 8-bit runs on the DGX Station GB300 with 1M of context at about 213 tokens a second.

Report VR-0C7A3DD, calculated Nov 22, 2026 from published specs, not benchmarked.