4× DGX Spark
87.7tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
Local AI · Recipe · 4× DGX Spark
DeepSeek-V4.1-Flash 552B EXL3 4.75 bpw with vLLM on 4× DGX Spark
by vcruz305
Intelligence (original model) · Artificial Analysis40with thinkingno thinking 25
Engine
- Engine
- vLLM
- Quantization
- EXL3 SAGE mixed-K 4.75bpw, Engram disk-backed FP8
- Model family
- DeepSeek-V4.1-Flash
- Context
- 1,048,576
- Parameters
- 552B Backbone + ~196B Engram, 384 Experten top-6 (lt. HF-Card)
- Creator
- vcruz305
- GitHub stars
- 13
- Repo updated
- Sep 25, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
No sourced measurements for this recipe.
What you need
- Hardware
- 4 × NVIDIA DGX Spark (GB10)
- Weights
- Main weightsvcruz305/DSV4.1-Flash-EXL3-4.75bpw MIT
- Basedeepseek-ai/DeepSeek-V4.1-Flash MIT
- Engine
- vLLM
- Context
- 1,048,576 tokens
Notes
What matters before you rebuild it.
- Multiple nodes
- Custom kernel required
- Measured by a third party
Sources
Related recipes
4× RTX PRO 6000
73.1tok/sEveryday
1 request, prose prompt, context 32788, with DSPARK Source
Intelligence40with thinkingno thinking 25
4× DGX Spark
59.1tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
3× DGX Spark
51tok/sEveryday
1 request, prose prompt, with MTP/EAGLE (spekulativ, ~59 ms/step) (per recipe) Source
Intelligence40with thinkingno thinking 25
- Custom kernel required
2× DGX Spark
31.6tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
2× DGX Spark
31.6tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
- ≤ 3 bit: quantization may cost quality
- Experimental
- Custom kernel required
Local AI in your company?
In a workshop we work out which models and which hardware fit your tasks, and build the first agent on your infrastructure.