In short
Renting a Tesla V100 Super 32 GB on gpu.nz: from $0.284 per hour for one GPU at the cheapest current offer, billed per minute from your balance. The price changes with availability, so the live figures are in the table below.
| GPU | Server | Region | Price $/h | Rent | |
|---|---|---|---|---|---|
Tesla V100 Super 32 GB983 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
Tesla V100 Super 32 GB983 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
2×Tesla V100 Super 32 GB983 GB/s | 20 vCPU · 157 GB RAM674 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
2×Tesla V100 Super 32 GB983 GB/s | 20 vCPU · 157 GB RAM674 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
4×Tesla V100 Super 32 GB983 GB/s | 40 vCPU · 315 GB RAM1,349 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
8×Tesla V100 Super 32 GB983 GB/s | 80 vCPU · 630 GB RAM2,699 GB · ↓ 3,989 | USUnited States | 95.4% | Rent | |
Tesla V100 Super 32 GB983 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
Tesla V100 Super 32 GB983 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
Tesla V100 Super 32 GB982 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
Tesla V100 Super 32 GB982 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
Tesla V100 Super 32 GB983 GB/s | 10 vCPU · 79 GB RAM337 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
2×Tesla V100 Super 32 GB983 GB/s | 20 vCPU · 157 GB RAM674 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
2×Tesla V100 Super 32 GB982 GB/s | 20 vCPU · 157 GB RAM674 GB · ↓ 3,989 | USUnited States | 95.4% | Busy | |
4×Tesla V100 Super 32 GB983 GB/s | 40 vCPU · 315 GB RAM1,345 GB · ↓ 3,989 | USUnited States | 95.4% | Busy |
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
20 vCPU · 157 RAM · 674 disk · net ↓ 3,989
95.4% · unverified
20 vCPU · 157 RAM · 674 disk · net ↓ 3,989
95.4% · unverified
40 vCPU · 315 RAM · 1,349 disk · net ↓ 3,989
95.4% · unverified
80 vCPU · 630 RAM · 2,699 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
10 vCPU · 79 RAM · 337 disk · net ↓ 3,989
95.4% · unverified
20 vCPU · 157 RAM · 674 disk · net ↓ 3,989
95.4% · unverified
20 vCPU · 157 RAM · 674 disk · net ↓ 3,989
95.4% · unverified
40 vCPU · 315 RAM · 1,345 disk · net ↓ 3,989
95.4% · unverified
Estimates for short context and typical settings. See the guide on memory for LLMs. How much VRAM an LLM needs: a table and a rule of thumb
| On-demand | per hour | per day |
|---|---|---|
| 1 GPU | $0.31 | $7.44 |
| 2 GPU | $0.596 | $14.30 |
| 4 GPU | $1.15 | $27.60 |
| 8 GPU | $2.27 | $54.48 |
Interruptible: $0.173/h per GPUInterruptible bid price per GPU. The server can be stopped if someone outbids you.
Prices and availability come from the live catalog and update automatically.
Countries: 1