Blended GPU compute costs rose 1.5% this week across major cloud providers, led by ALVEO_U30's 52.2% gain.
Top Movers
| GPU | Blended Price | WoW Change | Class |
|---|---|---|---|
| ALVEO_U30 | $0.06/hr | ▲ 52.2% | General |
| TPU_V5LITEPOD | $0.18/hr | ▲ 3.2% | General |
| B200 | $7.56/hr | ▲ 1.9% | Training |
| VIRTEX_VU9P | $0.94/hr | ▲ 1.6% | General |
| L40S | $1.33/hr | ▲ 0.7% | Inference |
| V100 32GB | $1.55/hr | ▼ 0.7% | Training |
| VIRTEX_VU47P | $0.52/hr | ▲ 0.7% | General |
Blended pricing = average of spot, on-demand, and 1-year reserved rates across major cloud providers.
Training vs Inference
Training-class GPU pricing held steady this week (avg $4.30/hr, +0.1% WoW), while inference-class pricing held steady (avg $0.52/hr, +0.1% WoW).
The training-to-inference price ratio stands at 8.3x — stable compared to last week. The elevated spread suggests strong demand for training compute relative to inference, consistent with ongoing large model training activity.
Regional Spotlight: Asia Pacific
Asia Pacific trades at a 18% premium to global averages this week, with 31 GPU types available across the region. The most expensive GPUs in the region are GB200 ($14.87/hr), H200 ($10.12/hr), TRAINIUM ($6.73/hr). The 18% regional premium reflects regional demand dynamics.
For detailed pricing data across all Asia Pacific sub-regions, see the full regional profile.
Implications
For cloud buyers: Europe continues to offer the lowest average GPU pricing ($2.23/hr blended average). For workloads with regional flexibility, the gap between Europe and Middle East is $1.85/hr — a 83% premium. Compare regional pricing →
For semiconductor analysts: GPU pricing trends remain broadly stable this week. H100 (+0.1% WoW) and MI300X (+0.0% WoW) are tracking within normal ranges. Blackwell (B200) blended pricing at $7.56/hr (+1.9% WoW) provides an early read on next-generation adoption curves. View all GPU profiles →
For GPU investors: Rising rental rates are improving returns for existing GPU deployments. Model scenarios with the GPU ROI Calculator →