Great work by @Inferact and @vllm_project on this amazing Rubin-optimized DeepSeek v4.1 Flash! https://x.com/vllm_project/status/2121008091538092393?a=lKcgDEqlTi4rA3zem8g&s=46 Quoted @vllm_project: We optimized DeepSeek-V4.1-Flash on vLLM for Vera Rubin NVL72 and reached 58× the throughput of GB300 NVL72 at 1.7 million tokens/s/GPU and 387 tokens/s/user on SemiAnalysis AgentX. Across the curve: 🔹 2.52 million tokens/s/GPU at peak throughput (2.7× GB300) 🔹 1.7 million tokens/s/GPU at P90 387 tokens/s/user (58× GB300) 🔹 1,810 tokens/s at one concurrent session with TP 1/4(Historical earnings: For 2026Q2 (period ended 2026-06-27), AMD reported basic EPS of 2.26 and diluted EPS of 2.22, net income of USD 3.680 billion, and revenue of USD 21.789 billion. Consensus expectations: For 2026Q3 (event date 2026-11-03), consensus EPS is 1.9366 with revenue estimate of USD 13.083 billion.)