Spark Bench

DeepSeek V4 Flash 0731 SparkInfer single-Spark serving baseline

1× DGX Spark · vLLM + SparkInfer ·

Tested configuration

Hardware
1x NVIDIA DGX Spark (128GB)
Serving runtime
vLLM + SparkInfer
Runtime version or build
vLLM with SparkInfer/B12X container ghcr.io/0xsero/deepseek-v4-flash-0731-spark-sparkinfer@sha256:2e077489a83a0360952828051fe7f7a32c1801e5ce8436d85f7267583d614ff4
Weight quantization or precision
EXL3
Benchmark date
Recipe credit
MiaAI-Lab DeepSeek V4 Flash single-Spark recipeBy MiaAI-Labd4ba142bc1d971eb73a911e207e3e963bbb3c455

Decode sweep

~256 input / 512 output tokens per request. Output throughput is aggregate across concurrent requests. TTFT means time to first token.

ConcurrencyOutput tok/sTTFT p95 (s)
129.620.54
231.0321.04
431.4454.29
831.08118.29
1229.40195.33
1629.60263.50

Prefill and context

Configured request window
384,000 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/sTTFT p95 (s)
4,096 input tokens653.493.95
32,768 input tokens1,046.8928.94
131,072 input tokens1,020.75126.73
245,760 input tokensPasskey prompt947.57259.00

Sustained load

20-minute soak at concurrency 8 · ~1024 input / 512 output tokens per request.

Aggregate throughput
26.89 output tok/s
TTFT p95
150.18 s
Completed requests
71
Request errors
0

Serving regression checks 13/13 passed

Tool-use evaluation

Unavailable · no published tool-eval result for this configuration.