Signal65: CoreWeave up to 65% cheaper on GB200 over 3 years, buys 2.9x the tokens
ryanshrout · x · 2026-10-01
A Signal65 analysis modeled a 5,000-GPU NVIDIA Blackwell deployment over three years on CoreWeave and three hyperscaler clouds. On price alone, CoreWeave came in up to 65% lower on GB200 NVL72 and 52% lower on HGX B300, with no egress, support, or observability fees. Throughput widens the gap: running DeepSeek-R1 on identical GB200 NVL72 hardware, CoreWeave posted the top per-GPU MLPerf Inference v6.0 result, beating NVIDIA's reference and delivering 2.2x the throughput of the only hyperscaler that submitted. That turns a 22% price advantage into 65% lower cost per million tokens — a $1M annual commitment buys 1.33 trillion tokens on CoreWeave vs 466 billion on the competitor. Three-year cost differences across providers run from hundreds of millions to over $2B.
More from Infra
- DRAM supply shows no line of sight to catching up with demand, says analyst — BenBajarin · 2026-10-01
- Micron: humanoid robots may need memory comparable to autonomous vehicles, driving demand by 2030 — McDonaghMatthew · 2026-10-01
- antirez praises 192GB Framework Desktop, plugs external GPU for heterogeneous inference — antirez · 2026-10-01
- RTX 5090 + Intel Arc B70 for local LLMs: halved throughput vs bigger context — indiealexh · 2026-10-01
- Google explores space data centers: keeping AI chips from overheating in orbit — GraceToSentience · 2026-10-01
- Magnitude inference engine hits #1 on HN, claims up to 2x faster local open-model runs than llama.cpp — nickbaumann_ · 2026-10-01