Llama 4 Scout 17B 16E Instruct on DGX Spark: VRAM, speed and fit
Llama 4 Scout 17B 16E Instruct at 8-bit runs on the DGX Spark 128 GB with 28K of context at about 9.7 tokens a second.
Text generation · Chat · Ollama / LM Studio
Runs comfortably · on DGX Spark 128 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat
- GPU memory
- 121 GB
- of 128 GB
- Context
- 28K
- auto
- Speed
- 9.7 tok/s
- 7.5 when full
- First reply
- 3.8 s
- 2K prompt
Where the memory goes
- Model
- 108 GB
- Context
- 5.3 GB
- Overhead
- 811 MB
- OS
- 7.7 GB
- Free
- 6.8 GB
- Memory
- 121 GB of 128 GB
- Context
- 28K tokens · auto
- Speed
- 9.7 tok/s · 7.5 when full
- First reply
- 3.8 s for a 2K prompt
- Running cost
- $3.15/mo at home · $287/mo on a cloud 4× A40
Validation: Validated with conditions
| Check | Target | Result | Status |
|---|---|---|---|
| Fits in GPU memory | 120 GB usable | 114 GB needed | Pass |
| Headroom | 5% free | 6.8 GB free (6%) | Pass |
| Context for chat | 16K or more | 28K | Pass |
| Writing speed | 15 tok/s or more | 9.7 tok/s | Check |
| First reply | under 3 s, 2K prompt | 3.8 s | Check |
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 4 Scout 17B 16E Instruct
- Parameters
- 109B, 18B active
- Precision
- 8-bit (Q8_0)
- Context
- 28K, auto
- Chats at once
- 1
- Use
- Chat
- Runs in
- Ollama / LM Studio
- Hardware
- DGX Spark 128 GB
- GPU can use
- 120 GB of 128 GB
Worth trying
- 8-bit cache: Room for 60K context
Running cost
- At home
- $3.15/mo
- 144 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 22, 2026.
How to run it
- llama.cpp
llama-server -hf unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF- vLLM
vllm serve meta-llama/Llama-4-Scout-17B-16E-Instruct
Llama 4 Scout 17B 16E Instruct on similar machines
- M5 Ultra 512 GBRuns
- Ryzen AI Max+ 395 64 GBRuns
- Ryzen AI Max+ 395 128 GBRuns
- RTX SparkRuns
- Dual DGX SparkRuns
- DGX Station GB300Runs
- 2× RTX PRO 6000 BlackwellRuns
- 4× RTX PRO 6000 BlackwellRuns
All machines: Llama 4 Scout 17B 16E Instruct VRAM requirements. Everything that runs here: what can the DGX Spark run?
Other models that run on DGX Spark
Questions
Can Llama 4 Scout 17B 16E Instruct run on DGX Spark?
Llama 4 Scout 17B 16E Instruct at 8-bit runs on the DGX Spark 128 GB with 28K of context at about 9.7 tokens a second.
Report VR-1US1E8J, calculated Nov 22, 2026 from published specs, not benchmarked.