4× RTX 3090
145tok/sEveryday
1 request, prose prompt, with DFlash (per recipe) Source
Intelligence34with thinkingno thinking 20
- Experimental
Local AI · Recipe · 1× DGX Spark
Qwen3.8 27B NVFP4 with vLLM on 1× DGX Spark (256k context)
Qwen3.8 27B NVFP4 with vLLM on 1× DGX Spark (256k context): 75.6 tok/s according to github.com (dataset as of Nov 3, 2026).
by vcruz305
Intelligence (original model) · Artificial Analysis34with thinkingno thinking 20
Engine
- Engine
- vLLM 0.1.dev1+g75231eff2 (Host-Build)
- Quantization
- NVFP4 (unsloth)
- Model family
- Qwen3.8
- Context
- 262,144
- Parameters
- 27B dense
- Creator
- vcruz305
- GitHub stars
- 0
- 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
What you need
- Hardware
- 1 × NVIDIA DGX Spark (GB10)
- Weights
- Main weightsunsloth/Qwen3.8-27B-NVFP4 Apache-2.0
- Draft modelDoopeworld/Qwen3.8-27B-DSpark-vLLM other
- othervcruz305/Qwen3.8-27B-GGUF Apache-2.0
- otherRadixArk/Qwen3.8-27B-NVFP4 Apache-2.0
- Engine
- vLLM 0.1.dev1+g75231eff2 (Host-Build)
- Context
- 262,144 tokens
Related recipes
2× RTX 3090
123tok/sEveryday
1 request, prose prompt, with Draft Source
Intelligence34with thinkingno thinking 20
4× RTX 3090
85.3tok/sEveryday
1 request, prose prompt, with DFlash2 (per recipe) Source
Intelligence34with thinkingno thinking 20
2× RTX 3090
76.4tok/sEveryday
1 request, prose prompt, with MTP Source
Intelligence34with thinkingno thinking 20
1× DGX Spark
58.3tok/sEveryday
1 request, prose prompt, with DFlash2 (per recipe) Source
Intelligence34with thinkingno thinking 20
- Custom kernel required
1× RTX 3090
57tok/sEveryday
1 request, prose prompt, with MTP Source
Intelligenceno independent value
- 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.