1d:I[70616,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"Breadcrumb"] 1e:I[90762,["/_next/static/chunks/0kcf5dt6yxdox.js","/_next/static/chunks/1su5n1uv1ti4s.js","/_next/static/chunks/0gvke1wv6udfg.js","/_next/static/chunks/1dfrle3jgmlzh.js","/_next/static/chunks/162lgvduyr-fa.js","/_next/static/chunks/2gcnnopam8e9w.js","/_next/static/chunks/2g60iab5a1pkg.js"],"ArticleGrid"] 1f:T1a9f,

An operator who cannot get another megawatt from the utility has only one lever left: extract more sellable output from the power already contracted. That is the argument NVIDIA built its AI Infra Summit keynote around this week, and the company came with a customer number to prove the lever moves — Lambda fit 19 servers into a power budget normally reserved for 16, and got 24% more tokens per second out of the cluster for it.

Key takeaways

Lambda's DSX MaxLPS deployment lifted cluster-wide token throughput from roughly 4 million to 5 million tokens per second and performance per watt by 23% — the first production validation of the software on Blackwell servers.
NVIDIA reported up to 30x higher throughput per megawatt and up to 45x lower cost per million tokens for Vera Rubin NVL72 against GB300 NVL72, measured on replayed agentic coding sessions rather than single requests.
A separate demonstration with Emerald AI had an AI facility answering hundreds of live demand signals from a municipal utility without disturbing priority jobs.

Why the yardstick is changing

Peak FLOPS made sense as a buying signal when silicon supply was the constraint. It makes much less sense in a building where the interconnection queue, not the purchase order, decides how much compute can run. NVIDIA's framing — set out by Ian Buck, its vice president of hyperscale and high-performance computing, in Tuesday's summit announcement — is that validated agentic tokens per megawatt is the figure that actually maps to revenue.

The benchmark community is moving the same way. SemiAnalysis built AgentX around recorded real-world agentic coding sessions, preserving context growth, tool-call latency and sub-agent spawning instead of timing isolated prompts. NVIDIA puts one such session at roughly 15 times the token volume of an ordinary chat request, with input and output lengths varying widely between calls. That is not a workload a per-request suite describes well, which is why NVIDIA positions AgentX as a complement to MLPerf inference testing rather than a replacement.

Against that yardstick, NVIDIA's published AgentX figures for Vera Rubin NVL72 running DeepSeek V4 Pro reach 30x the throughput per megawatt of the prior GB300 generation, with cost per million tokens down as much as 45x. Both are stated as ceilings, not typical results.

What the Lambda result shows in practice

DSX MaxLPS is the piece doing the work in Lambda's case. Rather than provisioning each rack for its worst-case draw and stranding whatever it does not use, the software tracks consumption across GPUs and racks continuously and redirects available headroom to the jobs that need it. Training and serving have different power profiles, so a mixed estate leaves a great deal of that headroom idle by default.

Three extra nodes inside an unchanged power envelope is a modest-sounding win that compounds at scale, and it needs no new substation, cabling or permit. NVIDIA expects the margin to widen on its next generation, projecting up to 40% additional GPU capacity and up to 35% higher token throughput for Vera Rubin NVL72 sites under the same megawatt ceiling. For latency-bound agent work, the company is also pairing Vera Rubin with Groq 3 LPX, reporting 2,529 output tokens per second per user on a 100K-context Qwen 3.8 27B run.

The grid becomes part of the architecture

The demonstration with the longest tail had little to do with chips. Through Silicon Valley Power's flexible-load interconnection programme, Emerald AI showed a facility cutting draw automatically in response to utility signals — hundreds of them — while critical jobs kept running. Emerald intends to build its Conductor power-management product on NVIDIA DSX Flex, which reads load-shedding requests, demand-response events and pricing changes and acts within a hierarchy the operator defines in advance.

The point of the exercise is regulatory as much as technical. Utilities ration interconnection capacity on the assumption that a data centre is an inflexible block of demand; a facility that can prove it will yield on request has a different case to make in the queue. That argument runs in parallel with the capital NVIDIA has been marshalling for AI factory buildouts, and it addresses the constraint money alone cannot clear.

How much to bank on the numbers

Every multiple here comes from a vendor or a partner, measured on workloads the vendor chose, and the headline comparisons pit a current generation against its immediate predecessor. Lambda's is the outlier worth weighting: it is a production cluster, the deltas are small enough to be plausible, and other operators running Blackwell can check it. The 30x and 45x AgentX claims sit on a public dashboard, which at least makes them contestable.

Partner news rounded out the slate — Amazon's Annapurna Labs is co-developing NVHBM custom memory, d-Matrix is attaching Raptor XPUs to Vera CPUs over NVLink Fusion, and Pinterest has Blackwell and Dynamo behind conversational visual search. With NVIDIA saying Vera Rubin is in full production, independent measurement should arrive soon enough to settle the rest.

FAQ

What is tokens per megawatt?

It is throughput normalised by electrical draw rather than by chip count or peak FLOPS. Where a utility connection rather than capital caps expansion, it tells an operator how much sellable output each contracted megawatt produces, which is closer to a revenue figure than a hardware spec.

How much did DSX MaxLPS improve Lambda's cluster?

Lambda ran 19 nodes in the power envelope usually allocated to 16 full-power nodes, raising cluster-wide token throughput roughly 24%, from about 4 million to 5 million tokens per second, and performance per watt by 23%. NVIDIA describes it as the first validation of MaxLPS on Blackwell servers.

Is the SemiAnalysis AgentX benchmark different from MLPerf?

Yes. AgentX replays whole agent trajectories, including tool calls and sub-agent spawning, while MLPerf Inference scores per-request performance across a wide range of models and scenarios. NVIDIA presents the two as complementary, since they answer different questions about the same hardware.

a:["$","div",null,{"className":"mx-auto max-w-7xl px-4 py-8 sm:px-6","children":[["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"BreadcrumbList\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https://ainewsway.com\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"#gpu\",\"item\":\"https://ainewsway.com/tag/gpu\"}]}"}}],["$","script",null,{"type":"application/ld+json","dangerouslySetInnerHTML":{"__html":"{\"@context\":\"https://schema.org\",\"@type\":\"ItemList\",\"@id\":\"https://ainewsway.com/tag/gpu#itemlist\",\"name\":\"#gpu Articles\",\"numberOfItems\":1,\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"url\":\"https://ainewsway.com/articles/nvidia-makes-tokens-per-megawatt-the-metric-and-lambda-puts-a-number-on-it-mu3n7jgs\",\"name\":\"NVIDIA Makes Tokens Per Megawatt the Metric, and Lambda Puts a Number on It\"}]}"}}],["$","$L1d",null,{"items":[{"label":"# gpu"}]}],["$","h1",null,{"className":"mb-8 text-3xl font-bold text-zinc-900 dark:text-zinc-100","children":["#","gpu"]}],[["$","$L1e",null,{"articles":[{"id":"61d24f39-d64f-4226-9371-6f5cb40dfbc4","slug":"nvidia-makes-tokens-per-megawatt-the-metric-and-lambda-puts-a-number-on-it-mu3n7jgs","title":"NVIDIA Makes Tokens Per Megawatt the Metric, and Lambda Puts a Number on It","subtitle":"First production validation of DSX MaxLPS squeezed 19 nodes into a 16-node power budget as agentic workloads rewrite the benchmark","content":"$1f","excerpt":"At AI Infra Summit, NVIDIA reframed AI infrastructure around tokens per megawatt, backed by Lambda's 24% throughput gain and new agentic benchmark results.","coverImage":"https://cdn.ainewsway.com/uploads/2026/09/5bda2c85-5939-42f5-8ea2-3569d5b2b9ff.webp","coverImageCaption":"NVIDIA is pitching AI data centres as power-constrained factories, where the figure that matters is how many tokens each megawatt produces.","authorId":"6773f67d-d7fa-4b19-8f38-384f10ac761b","status":"PUBLISHED","publishedAt":"2026-09-16T05:11:28.698Z","viewCount":0,"likeCount":0,"readingTime":5,"isFeatured":false,"isBreakingNews":false,"metaTitle":"NVIDIA Shifts AI Metrics to Tokens Per Megawatt","metaDescription":"NVIDIA's AI Infra Summit push: Lambda fit 19 nodes in a 16-node power budget with DSX MaxLPS, and Vera Rubin NVL72 hit 30x throughput per megawatt.","metaKeywords":["NVIDIA Vera Rubin NVL72","DSX MaxLPS","tokens per megawatt","AI Infra Summit 2026","SemiAnalysis AgentX","AI factory efficiency","Lambda Labs"],"ogImage":null,"aiSummary":"NVIDIA used the AI Infra Summit to reframe AI infrastructure performance around validated tokens per megawatt rather than peak throughput. Lambda supplied the first production validation of DSX MaxLPS on Blackwell servers, fitting 19 nodes into a 16-node power budget for a 24% gain in cluster token throughput. On the SemiAnalysis AgentX agentic benchmark, Vera Rubin NVL72 posted up to 30x higher throughput per megawatt than GB300 NVL72. Power, not silicon, is now the binding constraint.","aiTags":[],"sentiment":null,"articleType":"ANALYSIS","videoEmbed":null,"videoType":null,"seoScore":85,"isAutoGenerated":true,"isBackfill":false,"pipelineStage":null,"copyrightScore":null,"similarityScore":null,"matchedSourceId":null,"createdAt":"2026-09-16T05:11:01.710Z","updatedAt":"2026-09-16T09:12:47.251Z","contentUpdatedAt":"2026-09-16T09:12:47.250Z","deletedAt":null,"author":{"id":"6773f67d-d7fa-4b19-8f38-384f10ac761b","name":"Seung Jung","username":"Seung Jung","avatar":null},"categories":[{"id":"8126f4b3-ce4a-464f-bc70-7c3c39d5540b","name":"Tech & Business","slug":"tech-business"}],"tags":[{"id":"51075b6e-7339-4ca5-a00d-1b068653c310","name":"NVIDIA","slug":"nvidia"},{"id":"e9d9ed20-6211-4e4c-a2af-eb3c5eced7c7","name":"AI Infrastructure","slug":"ai-infrastructure"},{"id":"6321d27d-5afd-4988-8422-916b2cc0a65a","name":"Data Center","slug":"data-center"},{"id":"507069fd-a1de-4aaf-88e6-0d0ef9741dc0","name":"Vera Rubin","slug":"vera-rubin"},{"id":"3f49d736-93c2-461a-bc3c-524b15af8e14","name":"GPU","slug":"gpu"}],"translations":[{"id":"3601b429-a5bb-48ae-a002-06404e7b5f3a","language":"es","title":"NVIDIA convierte los tokens por megavatio en la métrica, y Lambda le pone cifra","subtitle":"La primera validación en producción de DSX MaxLPS metió 19 nodos en un presupuesto eléctrico de 16 mientras las cargas con agentes reescriben el benchmark","excerpt":"En el AI Infra Summit, NVIDIA replanteó la infraestructura de IA en torno a los tokens por megavatio, con el 24% más de rendimiento de Lambda como respaldo.","metaTitle":"NVIDIA mide la IA en tokens por megavatio","metaDescription":"NVIDIA mide la IA en tokens por megavatio: Lambda metió 19 nodos en un presupuesto eléctrico de 16 con DSX MaxLPS y Vera Rubin NVL72 rindió 30 veces más."},{"id":"531fdb67-14b0-4e5b-afba-d0f3cfea929b","language":"ja","title":"NVIDIA、指標を「メガワットあたりトークン」へ　Lambdaが最初の実測値を示す","subtitle":"DSX MaxLPSの初の本番検証で16ノード分の電力枠に19ノードを収めた。エージェント型ワークロードがベンチマークの前提を書き換えている","excerpt":"AI Infra SummitでNVIDIAがAIインフラの物差しをメガワットあたりトークンへ再定義した。Lambdaの24%の処理量向上と新しいエージェント型ベンチマークの結果が裏付けだ。","metaTitle":"NVIDIA、AI指標をメガワットあたりトークンへ転換","metaDescription":"NVIDIAがAI Infra Summitでメガワットあたりトークンを掲げた。LambdaはDSX MaxLPSで16ノード分の電力に19ノードを収め、Vera Rubin NVL72は30倍の処理量を記録した。"},{"id":"8074f48f-7fb5-4d74-be68-d3db8cf86eba","language":"ko","title":"NVIDIA, AI 성능 잣대를 MW당 토큰으로 바꾼다… Lambda가 첫 수치 내놨다","subtitle":"DSX MaxLPS 첫 상용 검증에서 16노드 전력 예산에 19노드를 밀어 넣었다. 에이전틱 워크로드가 벤치마크 기준을 다시 쓰고 있다","excerpt":"AI Infra Summit에서 NVIDIA가 AI 인프라 평가 기준을 MW당 토큰으로 재정의했다. Lambda의 24% 처리량 향상과 새 에이전틱 벤치마크 결과가 근거다.","metaTitle":"NVIDIA, AI 성능 잣대를 MW당 토큰으로 전환","metaDescription":"NVIDIA가 AI Infra Summit에서 MW당 토큰을 내세웠다. Lambda는 DSX MaxLPS로 16노드 전력에 19노드를 넣었고, Vera Rubin NVL72는 MW당 30배 처리량을 기록했다."},{"id":"51a75954-a3a0-4668-a3e1-6d19ae0169c4","language":"zh","title":"NVIDIA把「每兆瓦token」立為新指標，Lambda端出了第一個數字","subtitle":"DSX MaxLPS首次正式環境驗證，把19個節點塞進16個節點的電力額度；代理式負載正在改寫基準測試的規則","excerpt":"NVIDIA在AI Infra Summit上把AI基礎設施的衡量標準重新定義為每兆瓦token，並以Lambda的24%吞吐量增幅與新的代理式基準結果為證。","metaTitle":"NVIDIA把AI指標轉向每兆瓦token","metaDescription":"NVIDIA在AI Infra Summit主打每兆瓦token：Lambda以DSX MaxLPS把19節點塞進16節點電力額度，Vera Rubin NVL72每兆瓦吞吐量達30倍。"}],"celebrities":[{"id":"95f8f574-a1e4-4a77-8783-1419c1421b78","nameKo":"엔비디아","nameEn":"NVIDIA","slug":"nvidia","profileImage":null}]}]}],"$L20"]]}]

#gpu

NVIDIA Makes Tokens Per Megawatt the Metric, and Lambda Puts a Number on It

Subscribe to Newsletter

#gpu

NVIDIA Makes Tokens Per Megawatt the Metric, and Lambda Puts a Number on It