2× DGX Spark
22.3tok/sPeak
1 request, best value reported by source Source
Intelligence25with thinking
- Restrictive license
- License unclear
Local AI · Recipe · 2× DGX Spark
Hy3 295B-A21B NVFP4 with vLLM on 2× DGX Spark (with MTP)
Hy3 295B-A21B NVFP4 with vLLM on 2× DGX Spark (with MTP): 19.4 tok/s according to github.com (dataset as of Nov 3, 2026).
by drowzeys
Engine
- Engine
- vLLM
- Quantization
- NVFP4 compressed-tensors + turboquant_k8v4 KV (skip-layers 80)
- Model family
- Hy3
- Context
- 262,144
- Parameters
- 295B MoE / 21B active
- Creator
- drowzeys
- GitHub stars
- 1
- Repo updated
- Aug 17, 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
Time to first token
Shorter is better.
What you need
- Hardware
- 2 × NVIDIA DGX Spark (GB10)
- Weights
- Main weights0xSero/Hy3-299B-NVFP4 other
- Basetencent/Hy3 Apache-2.0
- Engine
- vLLM
- Context
- 262,144 tokens
Notes
What matters before you rebuild it.
- Restrictive license
- License unclear
Sources
Related recipes
2× DGX Spark
21.8tok/sPeak
1 request, best value reported by source Source
Intelligenceno independent value
2× DGX Spark
89.3tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligenceno independent value
- Custom kernel required
- License unclear
2× DGX Spark
74tok/sEveryday
1 request, prose prompt (long story), thinking off, with MTP Source
Intelligence40with thinking
- Custom kernel required
- License unclear
2× DGX Spark
69.6tok/sEveryday
1 request, prose prompt, with MTP (per recipe) Source
Intelligence40with thinking
- License unclear
2× DGX Spark
68tok/sEveryday
1 request, generation at 32K context (speed table, pair), thinking on, xhigh, with MTP3 (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.