LLM Inference Benchmark Explorer

Qwen3.5-397B-A17B on 4× DGX Spark — INT4, vLLM, TP4 inference benchmark

Open this configuration in the LLM Inference Benchmark Explorer

ModelQwen3.5-397B-A17B
Parameters397B
Intelligence Index21.4
Agentic Index—
Device4× DGX Spark
QuantizationINT4
Max C4
Chat Capacity16
Agentic Capacity6
TP4
DP—
PP—
EnginevLLM
Speculative Decoding—
CTTFT (ms)TPS (tok/s)Status
132538.14PASS
239330.54PASS
454124.54PASS
862218.83FAIL

In OpenZeka's measurement, Qwen3.5-397B-A17B (397B parameters), served in INT4 format with vLLM on 4× DGX Spark (TP=4), reached a generation speed of 38.1 tok/s per request and a time to first token (TTFT) of 325 ms with a single request.

Considering the speed targets and the available KV cache capacity, the estimated capacity is 16 users for chat use and 6 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.

Notes: Switch topology, TP=4. 2.24x faster than PP=3 at C1.

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