Qwen3.8-Flash-Next-NVFP4 MiaAI vLLM two-Spark 2200 MHz baseline
2× DGX Spark · vLLM ·
Tested configuration
- Hardware
- 2x NVIDIA DGX Spark (128GB each)
- Serving runtime
- vLLM
- Runtime version or build
- vLLM 0.1.dev20073+g8e685d198, MTP speculative decoding with 3 draft tokens · vllm/vllm-openai:qwen38-flash-next
- Weight quantization or precision
- NVFP4
- Benchmark date
- Recipe credit
- MiaAI-Lab Qwen dual-Spark recipeBy MiaAI-Lab
169fbad266f2791335a3102f0d3d625e7c295563
Decode sweep
~256 input / 512 output tokens per request. Output throughput is aggregate across concurrent requests. TTFT means time to first token.
| Concurrency | Output tok/s | TTFT p95 (s) |
|---|---|---|
| 1 | 47.51 | 0.36 |
| 2 | 81.83 | 0.83 |
| 4 | 135.45 | 1.25 |
| 8 | 213.72 | 1.18 |
| 12 | 210.19 | 19.07 |
| 16 | 210.03 | 20.76 |
Prefill and context
- Configured request window
- 1,000,000 tokens
- Longest tested prompt
- 245,760 target input tokens
- Validated retrieval
- ≈240K input tokens · passkey passed
Input throughput uses full request wall time. The passkey measures retrieval from one long prompt.
| Workload (c1) | Input tok/s | TTFT p95 (s) |
|---|---|---|
| 4,096 input tokens | 1,251.71 | 1.96 |
| 32,768 input tokens | 2,361.85 | 12.64 |
| 131,072 input tokens | 2,350.87 | 54.38 |
| 245,760 input tokensPasskey prompt | 2,146.46 | 113.18 |
Sustained load
20-minute soak at concurrency 8 · ~1024 input / 512 output tokens per request.
- Aggregate throughput
- 201.58 output tok/s
- TTFT p95
- 1.31 s
- Completed requests
- 477
- Request errors
- 0
Serving regression checks 13/13 passed
Tool-use evaluation
Unavailable · no published tool-eval result for this configuration.