LLM Inference Benchmark Explorer

DeepSeek V4 Flash 0731 on 8× DGX Spark — NVFP4, SGLang, TP4, DP2 inference benchmark

Open this configuration in the LLM Inference Benchmark Explorer

ModelDeepSeek V4 Flash 0731
Parameters304B
Intelligence Index34.3
Agentic Index41
Device8× DGX Spark
QuantizationNVFP4
Max C1+
Chat Capacity4+
Agentic Capacity1+
TP4
DP2
PP—
EngineSGLang
Speculative Decoding—
CTTFT (ms)TPS (tok/s)Status
123863.56PASS

In OpenZeka's measurement, DeepSeek V4 Flash 0731 (304B parameters), served in NVFP4 format with SGLang on 8× DGX Spark (TP=4, DP=2), reached a generation speed of 63.6 tok/s per request and a time to first token (TTFT) of 238 ms with a single request.

Considering the speed targets and the available KV cache capacity, the estimated capacity is at least 4 users for chat use and at least 1 for agentic use. The realistic capacity will likely fall between these two values. In scenarios dominated by coding, tool use, long workflows and multi-agent use, capacity approaches the agentic estimate; where shorter interactions, standard conversations and lighter tasks dominate, it approaches the chat estimate.

“At least” means the configuration met the speed targets even at the highest load tested, so the real capacity may be higher.

This configuration was measured with a single request only, so the estimate rests on one measurement.

Notes: TP=4, DP=2. Single C=1 test (scaling comparison)

Intelligence Index and Agentic Index values are published by Artificial Analysis and are reproduced here with attribution.