RTX 5000 Pro Runs Qwen3.8 at Unbeatable Value
Valuable-Run2129 · reddit · 2026-08-24
A user shared impressive benchmarks running Qwen3.8-27B on an RTX 5000 Pro (48GB VRAM): 130 t/s decode, 4000 t/s prefill, and 5 concurrent requests. Using SGLang, prefixes are offloaded to storage to prevent reprocessing in sub-agentic workflows. The author argues this model drastically increases local compute value, rivaling a Claude subscription and driving up the card's resale price.
More from Infra
- SK Hynix ties new US plant to looser chip equipment exports to China, sources say — pstAsiatech · 2026-10-15
- Perplexity's new embedding models: index with a 9B model, search with 0.6B — antoine_chaffin · 2026-10-15
- Unsloth ships GGUF quants for Nemotron 3.5: Bolt-74B runs in 24GB RAM — danielhanchen · 2026-10-15
- vLLM-Omni ships Day-0 serving for FLUX 3 Image Klein with BF16, FP8, NVFP4 support — vllm_project · 2026-10-15
- NVIDIA's Nemotron 3.5 Super VL lands training and inference support on Applied Compute — ypatil125 · 2026-10-15
- CoreWeave bond yield tops 16% as loans outlast customer contracts by ~2 years — tengyanAI · 2026-10-15