독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스주제 · CoreWeave动态중대2 분 소요DisconfirmAI 편집부

AI Cloud Firm CoreWeave Claims Leading DeepSeek-R1 Throughput in New MLPerf 0.7 Endpoints Benchmark

CoreWeave self-reported 441k tokens/sec system throughput for DeepSeek-R1 inference in the inaugural MLPerf Endpoints benchmark, emphasizing production-environment validation.

2026-10-23 00:27 GMT+8

US-based AI cloud infrastructure company CoreWeave announced on July 28, 2026, via its official blog that it achieved leading results in the inaugural MLPerf 0.7 Endpoints benchmark by MLCommons. Running the frontier DeepSeek-R1 reasoning model on a NVIDIA GB200 NVL72 system with 68 NVIDIA Blackwell GPUs, CoreWeave reported a sustained system throughput of 441,740 output tokens per second at 16,384 concurrent requests, and up to 6,496 tokens per second per GPU.

MLPerf Endpoints is MLCommons' newest benchmark suite, designed to measure how an inference service behaves under realistic, concurrent, online load, reporting the full throughput-versus-latency curve rather than a single data point. The test includes DeepSeek-R1, a 671-billion-parameter mixture-of-experts reasoning model that emits thousands of tokens per response, making it the hardest serving workload in the suite.

CoreWeave emphasized that its results were validated on standard production clusters using the same network fabric, scheduler, storage, and Mission Control orchestration that customers use, rather than an isolated environment tuned for the test. The company noted that throughput per GPU is the metric most directly reflecting the economics of real inference services, as it maps to tokens served per accelerator and cost per GPU.

It should be noted that these performance figures are self-reported by CoreWeave. While the submission process was verified by the MLCommons Association, independent third-party reproduction has not been publicly reported. Additionally, the 6,496 tokens per second per GPU figure is derived by dividing system TPS by the GPU count (68) and is not a direct MLPerf reported metric.

로그인 / 회원가입
독서리서치레이더투자 프레임워크

출처

  • wf.coreweave.com↗
주제

CoreWeave动态

美国AI云服务商CoreWeave在2026年10月密集发布技术产品并宣布全球扩张,但其高杠杆商业模式引发市场剧烈震荡。公司推出Forge平台及Nvidia Vera CPU支持,并在北达科他州和印度推进数据中心建设;然而,其债券收益率飙升…

7 단계15 건업데이트됨 2026-10-26 09:47

현재까지의 과정

  1. 2026-10-02CoreWeave宣布引入英伟达Vera CPU并上线对象存储跨区域写入功能;Applied Digital在北达科他州Polaris Forge 1园区新增75兆瓦算力,总规模达250兆瓦。wf.coreweave.com ↗
  2. 2026-10-03CoreWeave发布Forge平台,旨在连接推理、观察与模型改进闭环,支持模型检查点热加载。coreweave.com ↗
  3. 2026-10-07CoreWeave宣布进入印度市场,计划与AdaniConneX合作在孟买附近部署240兆瓦数据中心容量。investors.coreweave.com ↗

아직 해결되지 않은 문제

  • CoreWeave能否成功以更低利率进行再融资或通过资产出售避免流动性危机?
  • Meta等超大规模客户在本季度财报中是否维持原有的巨额AI资本支出计划?
  • 随着融资成本上升,是否有更多依赖债务的数据中心项目被推迟或取消?

이 스레드의 다른 내용

  • CoreWeave Blog: The True Driver of Agent Production Costs is Context Resend in Loops; Architecture Choices Matter More Than Model Selection2026-10-23
  • CoreWeave's Min Jun: Agentic Traffic Breaks Network Assumptions Never Tested by Training2026-10-23
전체 주제 읽기 →
리서치 계속하기 →