LLM Inference Benchmark Explorer

Gemma-4-31B-it on 4× DGX Spark — NVFP4, vLLM, TP4 inference benchmark

Open this configuration in the LLM Inference Benchmark Explorer

ModelGemma-4-31B-it
Parameters31B
Intelligence Index14.7
Agentic Index4.2
Device4× DGX Spark
QuantizationNVFP4
Max C16
Chat Capacity64
Agentic Capacity24
TP4
DP—
PP—
EnginevLLM
Speculative Decoding—
CTTFT (ms)TPS (tok/s)Status
128032.46PASS
224530.77PASS
430630.04PASS
837928.11PASS
1636821.95PASS
3248014.59FAIL
645658.17FAIL

In this measurement, Gemma-4-31B-it (31B parameters), served in NVFP4 format with vLLM on 4× DGX Spark (TP=4), reached a generation speed of 32.5 tok/s per request and a time to first token (TTFT) of 280 ms with a single request.

Considering the speed targets and the available KV cache capacity, the estimated capacity is 64 users for chat use and 24 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.

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