4× DGX Spark
87.7tok/sEveryday
1 request, prose prompt, with DSpark (per recipe) Source
Intelligence40with thinkingno thinking 25
Local AI · Recipe · 1× DGX Spark
DeepSeek-V4.1-Flash on 1× DGX Spark
by 0xBakeer
Intelligence (original model) · Artificial Analysis40with thinkingno thinking 25
≤ 3 bit: quantization may cost quality
Engine
- Engine
- eigene PyTorch-Engine + Triton-Kernel v0.5.0-wip
- Quantization
- FP4-Experts (Checkpoint) bzw. CB3 3-bit Codebook für resident set
- Model family
- DeepSeek-V4.1-Flash
- Context
- 262,144
- Parameters
- 510 GB on disk, 384 Experts/Layer, 6 aktiv
- Creator
- 0xBakeer
- GitHub stars
- 120
- Repo updated
- Sep 24, 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
- 1 × NVIDIA DGX Spark (GB10)
- Weights
- Main weightsdeepseek-ai/DeepSeek-V4.1-Flash MIT
- Engine
- eigene PyTorch-Engine + Triton-Kernel v0.5.0-wip
- Context
- 262,144 tokens
Notes
What matters before you rebuild it.
- ≤ 3 bit: quantization may cost qualityAt 3 bit and below the model may answer noticeably worse than the original. The intelligence number refers to the original.
- Experimental
- Work in progress
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.