NVIDIA H100 GPUs on CoreWeave’s AI Cloud Platform Achieve Record-Breaking Graph500 Run

Technology #AI Hardware 🏛️ Official|Analyzed: Jan 3, 2026 06:35•

Published: Dec 10, 2025 20:56

•

1 min read

Analysis

The article highlights a significant achievement in graph processing performance using NVIDIA H100 GPUs on CoreWeave's AI cloud platform. The record-breaking benchmark result of 410 trillion traversed edges per second (TEPS) demonstrates the power of accelerated computing for large-scale graph analysis. The focus is on the performance of a commercially available cluster, emphasizing accessibility and practical application.

Key Takeaways

•NVIDIA H100 GPUs on CoreWeave's AI cloud platform achieved a record-breaking Graph500 result.
•The system achieved 410 trillion traversed edges per second (TEPS).
•The performance was achieved on a commercially available cluster.

Reference / Citation

View Original

"NVIDIA announced a record-breaking benchmark result of 410 trillion traversed edges per second (TEPS), ranking No. 1 on the 31st Graph500 breadth-first search (BFS) list."

NVIDIA AIDec 10, 2025 20:56

* Cited for critical analysis under Article 32.

Older

OpenAI adds MCP support to Agents SDK

Newer

Proof of Fourier Extension Conjecture for Paraboloid