LLM Inference Benchmark Explorer

Qwen3.8-Flash-Next on DGX B300 — BF16, SGLang, TP2 inference benchmark

Open this configuration in the LLM Inference Benchmark Explorer

ModelQwen3.8-Flash-Next
Parameters176B
Intelligence Index39.8
Agentic Index53.6
DeviceDGX B300
QuantizationBF16
Max C64+
Chat Capacity256+
Agentic Capacity96+
TP2
DP—
PP—
EngineSGLang
Speculative Decoding—
CTTFT (ms)TPS (tok/s)Status
1108250.08PASS
2153181.91PASS
4183144.97PASS
8207104.12PASS
1637778.76PASS
3228765.39PASS
6433349.20PASS

In OpenZeka's measurement, Qwen3.8-Flash-Next (176B parameters), served in BF16 format with SGLang on DGX B300 (TP=2), reached a generation speed of 250.1 tok/s per request and a time to first token (TTFT) of 108 ms with a single request.

Considering the speed targets and the available KV cache capacity, the estimated capacity is at least 256 users for chat use and at least 96 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.

Notes: TP=2

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