Skip to main content
NVIDIA Vera Rubin Platform

Rubin is coming.
Your GPU strategy should be ready.

10x cheaper tokens. 4x fewer GPUs for training. The Vera Rubin platform ships H2 2026 and will reshape cloud GPU pricing. Here's what you need to know — and what to do right now.

10×
cheaper
Token Cost vs Blackwell
fewer
GPUs for MoE Training
288GB
HBM4
HBM per GPU
H2 2026
shipping
Availability

Per-GPU Comparison

Rubin vs Blackwell vs Hopper

SpecRubinBlackwell (B200)Hopper (H100)
FP4 Inference~50 PFLOPS~9 PFLOPSN/A
FP8 Compute~25 PFLOPS~9 PFLOPS~1.98 PFLOPS
HBM Capacity288 GB HBM4192 GB HBM3e80 GB HBM3
Memory Bandwidth22 TB/s8 TB/s3.35 TB/s
NVLink Bandwidth3.6 TB/s1.8 TB/s900 GB/s
Process NodeTSMC 3 nmTSMC 4 nmTSMC 4 nm
Transistors~336B~208B~80B
Token Cost (est.)~$0.005/1M~$0.05/1M~$0.20/1M

Rack-Scale Comparison

NVL72 Rack: Rubin vs Blackwell

MetricVera Rubin NVL72Blackwell NVL72
Total GPUs72 Rubin72 B200
FP4 Inference3.6 EFLOPS~720 PFLOPS
HBM Capacity20.7 TB13.8 TB
HBM Bandwidth~1.58 PB/s~576 TB/s
NVLink Rack BW260 TB/s130 TB/s
Power Draw~120–130 kW~120 kW

Cloud Availability

Who's getting Rubin and when

Rubin entered full production January 2026. Volume is constrained to 200K–300K GPUs by TSMC 3nm and HBM4 supply.

AWSH2 2026

1M+ NVIDIA GPUs (Blackwell + Rubin)

Oracle OCIH2 2026

Rubin NVL72 Superclusters, 17+ ZetaFLOPS

AzureH2 2026

Rubin NVL72 in Fairwater AI superfactories

Google CloudH2 2026

Rubin deployments confirmed

CoreWeaveH2 2026

10x token throughput vs Blackwell

Lambda LabsH2 2026

Rubin NVL72 in Superintelligence Cloud

Nebius2027

Rubin deployments announced

Nscale2027

Rubin deployments announced

Decision Tool

Should you wait for Rubin?

Answer three questions. We'll give you a concrete recommendation.

1. What's your primary workload?

2. When do you need compute?

3. Budget sensitivity?

Available Now

Blackwell & Hopper: what you can rent today

While you wait for Rubin, these are the cheapest Blackwell and Hopper instances across our tracked providers.

GPUProviderVRAMGenPrice/hr
5060
RTX5060Ti
Vast.ai16GBBlackwell$0.0360/hrDeploy
5060
RTX5060Ti
Vast.ai16GBBlackwell$0.0497/hrDeploy
5070
RTX5070Ti
Vast.ai16GBBlackwell$0.0640/hrDeploy
5060
RTX5060Ti
Vast.ai16GBBlackwell$0.0667/hrDeploy
5060
RTX5060Ti
Vast.ai16GBBlackwell$0.0667/hrDeploy
5060
RTX5060Ti
Vast.ai16GBBlackwell$0.0693/hrDeploy
5070
RTX5070
Vast.ai12GBBlackwell$0.0773/hrDeploy
5090
RTX5090
Vast.ai32GBBlackwell$0.100/hrDeploy

Track the Rubin rollout with us

We'll add Rubin instances to our tracker as soon as cloud providers launch them. Check back or read our deep-dive article for the full GTC 2026 breakdown.

We use cookies for analytics and to remember your preferences. Privacy Policy