GPU.ai vs Runcrate
As of , GPU.ai lists the lower on-demand rate on 13 of the 18 GPUs both offer on-demand; Runcrate wins on 5. Not counted: 2 more shared GPUs, on community capacity or a marketplace floor. On the priciest shared card both list on-demand, NVIDIA B300: GPU.ai $10.79 vs Runcrate $17.02 per GPU-hour on-demand — GPU.ai is $6.23 (37%) cheaper.
On-demand, per GPU-hour
| GPU |  GPU.ai |  Runcrate | Difference | Cheaper |
|---|---|---|---|---|
| NVIDIA B300 median $9.00 · 26 providers · lowest Deep Infra $5.89 | $10.79 from captured Deploy ↗ | $17.02 from captured Deploy ↗ | $6.23 (37%) | GPU.ai |
| NVIDIA B200 median $7.17 · 37 providers · lowest Hostinger $4.50 from | $9.59 from captured Deploy ↗ | $9.33 from captured Deploy ↗ | $0.26 (3%) | Runcrate |
| NVIDIA H200 median $4.54 · 53 providers · lowest Dataoorts $2.29 | $4.31 from captured Deploy ↗ | $5.13 from captured Deploy ↗ | $0.82 (16%) | GPU.ai |
| NVIDIA H200 NVL median $3.87 · 8 providers · lowest IONOS Cloud $3.32 | $3.95 from captured Deploy ↗ | $4.67 from captured Deploy ↗ | $0.72 (15%) | GPU.ai |
| NVIDIA RTX PRO 6000 Blackwell median $2.19 · 42 providers · lowest Hostinger $0.65 from | $2.03 from captured Deploy ↗ | $2.28 from captured Deploy ↗ | $0.25 (11%) | GPU.ai |
| NVIDIA RTX 5090 median $0.76 · 24 providers · lowest OpenRelay $0.52 | $1.67 from captured Deploy ↗ | $0.72 from captured Deploy ↗ | $0.95 (57%) | Runcrate |
| NVIDIA A100 80GB median $2.04 · 60 providers · lowest Yotta Labs $0.92 | $1.48 from captured Deploy ↗ | $1.73 from captured Deploy ↗ | $0.25 (14%) | GPU.ai |
| NVIDIA RTX 4090 median $0.58 · 17 providers · lowest OpenRelay $0.35 | $1.07 from captured Deploy ↗ | $0.58 from captured Deploy ↗ | $0.49 (46%) | Runcrate |
| NVIDIA L40S median $1.29 · 39 providers · lowest CloudRift $0.69 | $0.96 from captured Deploy ↗ | $0.85 from captured Deploy ↗ | $0.11 (11%) | Runcrate |
| NVIDIA L40 median $0.98 · 12 providers · lowest Thunder Compute $0.79 | $0.94 from captured Deploy ↗ | $1.26 from captured Deploy ↗ | $0.32 (25%) | GPU.ai |
| NVIDIA L4 median $0.80 · 22 providers · lowest Hikube $0.27 GPU only | $0.83 from captured Deploy ↗ | $1.36 from captured Deploy ↗ | $0.53 (39%) | GPU.ai |
| NVIDIA A100 40GB median $1.98 · 21 providers · lowest Trooper.AI $0.67 from | $0.80 from captured Deploy ↗ | $1.99 from captured Deploy ↗ | — | not the same market |
| NVIDIA RTX 6000 Ada median $1.09 · 13 providers · lowest GPU.ai $0.78 from | $0.78 from captured Deploy ↗ | $1.20 from captured Deploy ↗ | $0.42 (35%) | GPU.ai |
| NVIDIA A40 median $0.95 · 9 providers · lowest CloudGPU $0.39 | $0.71 from captured Deploy ↗ | $2.66 from captured Deploy ↗ | $1.95 (73%) | GPU.ai |
| NVIDIA GeForce RTX 3090 median $0.42 · 11 providers · lowest OpenRelay $0.18 | $0.60 from captured Deploy ↗ | $0.41 from captured Deploy ↗ | $0.19 (32%) | Runcrate |
| NVIDIA RTX A6000 median $0.59 · 22 providers · lowest Thunder Compute $0.35 | $0.54 from captured Deploy ↗ | $0.57 from captured Deploy ↗ | $0.03 (5%) | GPU.ai |
| NVIDIA RTX 4000 Ada median $0.64 · 6 providers · lowest RunPod $0.28 | $0.34 from captured Deploy ↗ | $1.13 from captured Deploy ↗ | $0.79 (70%) | GPU.ai |
| NVIDIA RTX A5000 median $0.70 · 12 providers · lowest RunPod $0.27 | $0.32 from captured Deploy ↗ | $2.04 from captured Deploy ↗ | $1.72 (84%) | GPU.ai |
| NVIDIA RTX A4000 median $0.22 · 12 providers · lowest GPU.ai $0.14 from | $0.14 from captured Deploy ↗ | $0.22 from captured Deploy ↗ | $0.08 (36%) | GPU.ai |
| NVIDIA V100 median $0.79 · 13 providers · lowest Trooper.AI $0.24 from | $0.14 from captured Deploy ↗ | $0.36 from captured Deploy ↗ | — | not the same market |
How this is compared
Cloud GPU rental prices of GPU.ai and Runcrate, compared per GPU-hour on the GPUs both list. Where a provider has several tiers or regions, its cheapest on-demand rate is used and labelled. Community capacity (RunPod Community Cloud) and marketplace floors are shown only where a provider has nothing else, and never decide which is cheaper.
GPU by GPU
- NVIDIA B300: GPU.ai $10.79 vs Runcrate $17.02 per GPU-hour on-demand — GPU.ai is $6.23 (37%) cheaper.
- NVIDIA B200: GPU.ai $9.59 vs Runcrate $9.33 per GPU-hour on-demand — Runcrate is $0.26 (3%) cheaper.
- NVIDIA H200: GPU.ai $4.31 vs Runcrate $5.13 per GPU-hour on-demand — GPU.ai is $0.82 (16%) cheaper.
- NVIDIA H200 NVL: GPU.ai $3.95 vs Runcrate $4.67 per GPU-hour on-demand — GPU.ai is $0.72 (15%) cheaper.
- NVIDIA RTX PRO 6000 Blackwell: GPU.ai $2.03 vs Runcrate $2.28 per GPU-hour on-demand — GPU.ai is $0.25 (11%) cheaper.
- NVIDIA RTX 5090: GPU.ai $1.67 vs Runcrate $0.72 per GPU-hour on-demand — Runcrate is $0.95 (57%) cheaper.
- NVIDIA A100 80GB: GPU.ai $1.48 vs Runcrate $1.73 per GPU-hour on-demand — GPU.ai is $0.25 (14%) cheaper.
- NVIDIA RTX 4090: GPU.ai $1.07 vs Runcrate $0.58 per GPU-hour on-demand — Runcrate is $0.49 (46%) cheaper.
- NVIDIA L40S: GPU.ai $0.96 vs Runcrate $0.85 per GPU-hour on-demand — Runcrate is $0.11 (11%) cheaper.
- NVIDIA L40: GPU.ai $0.94 vs Runcrate $1.26 per GPU-hour on-demand — GPU.ai is $0.32 (25%) cheaper.
- NVIDIA L4: GPU.ai $0.83 vs Runcrate $1.36 per GPU-hour on-demand — GPU.ai is $0.53 (39%) cheaper.
- NVIDIA A100 40GB: GPU.ai $0.80 (from) vs Runcrate $1.99 (from) per GPU-hour — not the same market, not counted.
- NVIDIA RTX 6000 Ada: GPU.ai $0.78 vs Runcrate $1.20 per GPU-hour on-demand — GPU.ai is $0.42 (35%) cheaper.
- NVIDIA A40: GPU.ai $0.71 vs Runcrate $2.66 per GPU-hour on-demand — GPU.ai is $1.95 (73%) cheaper.
- NVIDIA GeForce RTX 3090: GPU.ai $0.60 vs Runcrate $0.41 per GPU-hour on-demand — Runcrate is $0.19 (32%) cheaper.
- NVIDIA RTX A6000: GPU.ai $0.54 vs Runcrate $0.57 per GPU-hour on-demand — GPU.ai is $0.03 (5%) cheaper.
- NVIDIA RTX 4000 Ada: GPU.ai $0.34 vs Runcrate $1.13 per GPU-hour on-demand — GPU.ai is $0.79 (70%) cheaper.
- NVIDIA RTX A5000: GPU.ai $0.32 vs Runcrate $2.04 per GPU-hour on-demand — GPU.ai is $1.72 (84%) cheaper.
- NVIDIA RTX A4000: GPU.ai $0.14 vs Runcrate $0.22 per GPU-hour on-demand — GPU.ai is $0.08 (36%) cheaper.
- NVIDIA V100: GPU.ai $0.14 (from) vs Runcrate $0.36 (from) per GPU-hour — not the same market, not counted.
Not compared: GPUs only one of the two lists, reserved/committed tiers, and anything either provider prices only on request. See each provider's page for its full list.
Data updated: