1× DGX Spark
71.8tok/sEveryday
1 request, prose prompt (sampled, thinking off), with LiLiCorr (per recipe) Source
Intelligenceno independent value
- ≤ 3 bit: quantization may cost quality
- Custom kernel required
- License unclear
Local AI · Recipe · 1× DGX Spark
Gemma-4-31B NVFP4 with vLLM on 1× DGX Spark
Gemma-4-31B NVFP4 with vLLM on 1× DGX Spark: 6.88 tok/s according to github.com (dataset as of Nov 3, 2026).
by r0b0tlab
Intelligence (original model) · Artificial Analysis19with thinkingno thinking 14
Engine
- Engine
- vLLM 0.21.0
- Quantization
- NVFP4 (ModelOpt) + FP8 KV
- Model family
- Gemma-4-31B
- Context
- 8,192
- Parameters
- 31B dense
- Creator
- r0b0tlab
- GitHub stars
- 4
- Repo updated
- May 19, 2026
Measurements
Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.
Decode · 1 request
What you need
- Hardware
- 1 × NVIDIA DGX Spark (GB10)
- Weights
- Main weightsnvidia/Gemma-4-31B-IT-NVFP4 other
- Engine
- vLLM 0.21.0
- Context
- 8,192 tokens
Related recipes
1× DGX Spark
60.3tok/sEveryday
1 request, prose prompt (essay), context 2k, no drafter (per recipe) Source
Intelligenceno independent value
- Custom kernel required
1× DGX Spark
59.4tok/sEveryday
1 request, chat prompt (20 prompts, thinking off), no drafting (default profile) Source
Intelligenceno independent value
- Custom kernel required
1× DGX Spark
58.3tok/sEveryday
1 request, prose prompt, with DFlash2 (per recipe) Source
Intelligence34with thinkingno thinking 20
- Custom kernel required
1× DGX Spark
55.6tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence40with thinking
- Custom kernel required
- License unclear
1× DGX Spark
53.3tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence40with thinking
- Custom kernel required
- License unclear
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.