Vera Rubin NVL72 goes live on CoreWeave as Cognition sees 4.8x token throughput vs GB200
ayushthakur0 · x · 2026-10-01
At CoreWeave Fully Connected, NVIDIA and CoreWeave announced production availability of Vera Rubin NVL72 systems with Spectrum-X 102.4T Ethernet. Cognition, the lab behind Devin, is the first customer running production workloads on the new systems, measuring 4.8x higher token throughput vs. GB200 NVL72 on real-world software engineering workloads spanning training, RL, and inference. CoreWeave will also offer Vera, NVIDIA's first CPU built for AI agents, and launched CoreWeave Forge, a connected environment for training, evaluating, and improving models and agents. NVIDIA highlighted platform durability: V100 GPUs from Volta still serve customer workloads nearly a decade after launch.
More from Infra
- Sarvam AI Launches Token Factory: India-Hosted Inference for Open Models — itsOmSarraf_ · 2026-11-16
- US BIS Closes Remote-Access Loophole for Advanced AI Chips in New Rule — apples_jimmy · 2026-11-16
- Jailbroken PS4 runs a 9B open-source model at up to 43 tok/s by talking directly to its 2013-era GPU — AIFlow_ML · 2026-11-16
- Mobile AI Factory: agents auto-build on-device SLMs with 19x less GPU time — pmttyji · 2026-11-16
- Stratechery's Fictional Obituary: Imagining Jensen Huang's 2026 Exit From Nvidia — Gregory_C_Allen · 2026-11-16
- vLLM serves DeepSeek V4.1-Flash at 954K tokens/s per GPU, $38.96/GPU-hour on GB300 NVL72 — bookwormengr · 2026-11-16