Qwen3.8 Flash hits 415 tok/s on dual DGX Sparks
NVIDIAAI · x · 2026-09-01
Benchmarked Qwen3.8 Flash Next on 2 DGX Sparks using NVFP4 format with 64 concurrent users. It generated 32,768 output tokens in 78.82 seconds, achieving a throughput of 415.7 tok/s. The test passed a 64K/user usable-context stress test with zero leakage, noted as a limit test rather than a production recipe.
More from Infra
- SGLang Summit sponsors grow from 11 to 36 in one week; Lip-Bu Tan, Lilian Weng to speak — BanghuaZ · 2026-10-18
- Yacine jokes his datacenter KPI is gigawatt-hours: get paged if not burning enough power — yacineMTB · 2026-10-18
- 1.68x inference speedup drops, directly accelerating GRPO post-training — yacineMTB · 2026-10-18
- OpenRouter is only a fraction of the total addressable market, poster argues — sloppenheimer · 2026-10-18
- Developer builds change alerts for LLM prices and retirements across Bedrock, Azure and Vertex — jsandhu0780 · 2026-10-18
- 12.3M tokens in 27 minutes for $0.14: agent session with 99.6% cache hit — teortaxesTex · 2026-10-18