{"slug":"vera-rubin","label":"Vera Rubin","item_count":3,"day_count":2,"source_count":2,"first_seen":"2026-11-12T00:00:00+00:00","last_updated":"2026-11-18T06:33:45+00:00","generated_at":"2026-11-19T00:13:28.317875+00:00","sources":["search_cn_open_weight_labs","vllm_blog"],"days":[{"date":"2026-11-12","items":[{"title":"H3-Image Day-0 Support: 4K in 0.87s with Vera Rubin","url":"https://vllm.ai/blog/2026-11-12-minimax-h3-image-rubin","source":"vllm_blog","type":"news","summary_1line":"H3-Image on Vera Rubin: 0.87-second 4K generation or 6,115 images/hour on four GPUs, with projected API revenue and support across consumer and data-center hardware.","sid":"66554ba00af6e4cb","published":"2026-11-12T00:00:00+00:00","editor_note":"First Rubin data point: vLLM's day-0 MiniMax H3-Image support, 4K in 0.87s or 6,115 images/hour on four GPUs."},{"title":"DeepSeek-V4.1-Flash on Vera Rubin NVL72: 58x Throughput vs. GB300 NVL72 at 387 tokens/s","url":"https://vllm.ai/blog/2026-11-12-deepseek-v41-flash-vera-rubin","source":"vllm_blog","type":"news","summary_1line":"On one NVIDIA Vera Rubin NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 1.7M tokens/s per GPU at a P90 of 387 tokens/s per user, 58x GB300 NVL72 at the same interactivity.","why_it_matters":"Matches feed focus: agent.","sid":"33948c421834d8b2","published":"2026-11-12T00:00:00+00:00","editor_note":"Source of the 58x claim: DeepSeek-V4.1-Flash on one NVL72 rack vs GB300 NVL72 at a P90 of 387 tokens/s per user."}]},{"date":"2026-11-18","items":[{"title":"CoreWeave Reports 2.7x Speedup with Vera Rubin Over GB200 on MLPerf DeepSeek-V3 Benchmark","url":"https://news.google.com/rss/articles/CBMizAFBVV95cUxNUS1oa25kdERDNUVjb0tGck44SnNhZGM5eTlmOV9tbmRyRk4yLTZHQ1RlVk9raDJsS1ZsZkNRLVY0RXJGMDA4Ni1iVGRlVi1nN0xyZ3JmYjBJRTJFMmFkbHA0RzJIQVlPeGNMLWp6Rm02QkNGUjhoaG9ZV2hTXzhWTzVQSFN4NS1YTmxVOXdDNnNkdzItbnZQenFIZktBNHV0SEh6MFpOSWtfN1lmbVZoLW85Q2w4ZlBuSG8ydE0tbTFxcWVMOWtaMlZDT2o?oc=5","source":"search_cn_open_weight_labs","type":"news","summary_1line":"CoreWeave Reports 2.7x Speedup with Vera Rubin Over GB200 on MLPerf DeepSeek-V3 Benchmark HPCwire","sid":"8f051756c4ba2fbd","published":"2026-11-18T06:33:45+00:00","editor_note":"First standardized-benchmark figure: CoreWeave's MLPerf DeepSeek-V3 run at 2.7x over GB200, reported by HPCwire."}]}],"editorial":{"tldr":"This is a run of separate vendor benchmark posts for NVIDIA's Vera Rubin, not one event. On Nov 12, vLLM published two day-0 serving results, for MiniMax H3-Image and DeepSeek-V4.1-Flash on a Vera Rubin NVL72 rack.","stale":false,"whats_new":"CoreWeave reports a 2.7x speedup for Vera Rubin over GB200 on the MLPerf DeepSeek-V3 benchmark (via HPCwire), the first Rubin number in this thread on a standardized suite.","why_it_matters":"Rubin speedup claims range from 2.7x on MLPerf to 58x on a vendor agent workload, so the generation-over-generation gain you plan capacity and cost around depends on which baseline and interactivity target you trust.","take_for_builders":"Hold Rubin capacity and cost projections until you rerun your own workload at your target per-user tokens/s; treat 58x as a vendor ceiling and 2.7x on MLPerf as the more conservative planning number.","beats":[{"kicker":"DAY-0 SERVING","tone":"launch","headline":"vLLM posts day-0 Rubin results for two open-weight models","summary":"Two blog posts from the vLLM project: image generation on four GPUs and DeepSeek-V4.1-Flash on a full NVL72 rack, with the headline 58x figure measured against GB300 at matched per-user speed.","sids":["66554ba00af6e4cb","33948c421834d8b2"]},{"kicker":"STANDARD BENCHMARK","tone":"now","headline":"CoreWeave's MLPerf run puts Rubin at 2.7x over GB200","summary":"A cloud provider's result on a standardized suite, with a far smaller multiplier than the vLLM agent-workload figure.","sids":["8f051756c4ba2fbd"]}],"open_questions":["Does MLCommons publish an audited MLPerf result that matches CoreWeave's 2.7x?","Does any third party reproduce vLLM's 58x DeepSeek-V4.1-Flash figure against GB300 NVL72?","When do CoreWeave or other clouds list Vera Rubin instances with public per-hour pricing?"],"generated_at":"2026-11-19T00:12:36Z"}}