NVIDIA Tesla V100 cloud GPU rental
Rent an NVIDIA Tesla V100 cloud GPU from $0.206 per hour with 16 GB of HBM2 at 900 GB/s of memory bandwidth. FastInfra bills pods per minute with no commitment, so a one-hour experiment costs $0.206 and you stop paying the moment you delete the pod. Run 1 to 8 GPUs per pod, connect over SSH or JupyterLab, and see the exact price before you deploy.
Prices synced November 3, 2026 · billed per minute
Tesla V100 specs and pricing
The NVIDIA Tesla V100 is a datacenter Volta GPU with 16 GB of HBM2 at 900 GB/s of memory bandwidth. On FastInfra it is a good fit for small-model inference, embeddings, prototyping and CI workloads. As a rule of thumb, one Tesla V100 holds the weights of a model of about 6B parameters at 16-bit precision, or about 25B with 4-bit quantization, after leaving room for the KV cache.
The Tesla V100 runs on the Value tier, lower-cost community hosts, at $0.206 per GPU-hour.
Tesla V100 price per hour by GPU count
Live FastInfra prices per pod-hour. Pods bill per minute, so you pay only for the minutes a pod exists.
| Pod size | Total VRAM | Value / hr | Month (730 h, lowest) |
|---|---|---|---|
| 1× Tesla V100 | 16 GB | $0.206 | $150 |
| 2× Tesla V100 | 32 GB | $0.412 | $301 |
| 4× Tesla V100 | 64 GB | $0.824 | $602 |
| 8× Tesla V100 | 128 GB | $1.65 | $1,203 |
What a Tesla V100 costs over time
One GPU at the lowest on-demand rate ($0.206/hr).
| Duration | On-demand |
|---|---|
| 1 hour | $0.206 |
| 8-hour working day | $1.65 |
| 24 hours | $4.94 |
| 1 week (168 hours) | $34.61 |
| 1 month (730 hours) | $150 |
Which LLMs fit on the Tesla V100?
GPUs needed to hold the model weights with about 20% of VRAM left for the KV cache. Rule of thumb only: long contexts and big batches need more.
| Model size | Examples | FP16 / BF16 | 8-bit | 4-bit |
|---|---|---|---|---|
| 7–8B | Llama 3.1 8B, Qwen3 8B, Mistral 7B | 2 GPUs | 1 GPU | 1 GPU |
| 12–14B | Gemma 3 12B, Qwen3 14B, Phi-4 | 4 GPUs | 2 GPUs | 1 GPU |
| 27–32B | Gemma 3 27B, Qwen3 32B | 8 GPUs | 4 GPUs | 2 GPUs |
| 70B | Llama 3.3 70B | Needs more GPUs | 8 GPUs | 4 GPUs |
| 120B | GPT-OSS 120B | Needs more GPUs | Needs more GPUs | 8 GPUs |
| 235B | Qwen3 235B-A22B | Needs more GPUs | Needs more GPUs | Needs more GPUs |
Best for
Small-model inference, embeddings, prototyping and CI workloads.
About the Volta architecture
Volta was the first architecture with Tensor Cores. It supports FP16 but not BF16, and libraries such as FlashAttention 2 require newer GPUs, so modern LLM code may need FP16 or slower fallback kernels. It remains a low-cost choice for classic deep learning, scientific computing and CUDA development.
Only need inference?
If you want to call an open model such as Llama, Qwen, DeepSeek or GPT-OSS rather than manage a server, the pay-per-token inference API is often cheaper than keeping a GPU busy around the clock. It is OpenAI-compatible and uses the same account and wallet as your pods. Browse every model with live prices.
Tesla V100 vs similar GPUs
Live lowest on-demand price per GPU-hour, and price per GB of VRAM.
| GPU | VRAM | Architecture | From / hr | Per GB-hr |
|---|---|---|---|---|
| Tesla V100 (this page) | 16 GB | Volta | $0.206 | $0.013 |
| NVIDIA RTX A4000 | 16 GB | Ampere | $0.184 | $0.012 |
| NVIDIA V100 SXM2 | 16 GB | Volta | $0.249 | $0.016 |
| NVIDIA RTX 2000 Ada | 16 GB | Ada Lovelace | $0.26 | $0.016 |
| NVIDIA RTX 4080 | 16 GB | Ada Lovelace | $0.292 | $0.018 |
| NVIDIA RTX 4080 SUPER | 16 GB | Ada Lovelace | $0.303 | $0.019 |
| NVIDIA RTX 5080 | 16 GB | Blackwell | $0.422 | $0.026 |
Tesla V100 rental — frequently asked questions
How much does it cost to rent an NVIDIA Tesla V100 per hour?
$0.206 per GPU-hour on demand. Billing is per minute while the pod exists, so 24/7 use for a 730-hour month costs about $150 per GPU at the lowest on-demand rate. Prices sync from live capacity and are shown before you deploy.
How much VRAM does the NVIDIA Tesla V100 have?
16 GB of HBM2 at 900 GB/s of memory bandwidth. The chip is rated at up to 250 W. A pod with 8 GPUs gives 128 GB of combined VRAM for tensor- or pipeline-parallel workloads.
What size LLM can I run on an NVIDIA Tesla V100?
Roughly 6B parameters at FP16 or BF16, 12B at 8-bit and 25B at 4-bit on a single GPU, reserving about 20% of VRAM for the KV cache and activations. Mixture-of-experts models need memory for all experts, not only the active ones. Long contexts and large batches need more headroom.
Is the NVIDIA Tesla V100 billed per hour or per minute?
Per minute. The hourly rate is divided by 60 and charged for every minute the pod exists. Attached volumes bill per GB-month and continue while a pod is stopped; delete the pod to stop all charges. There is no minimum term or reservation fee.
How do I rent an NVIDIA Tesla V100 on FastInfra?
Create an account, add prepaid credit, open the GPU configurator and choose the Tesla V100, the tier, the GPU count, a container template and storage. The configurator shows the hourly and monthly price, then deploys a pod you reach over SSH or JupyterLab within minutes.
Can I rent multiple Tesla V100 GPUs in one pod?
Yes, up to 8 per pod. The price scales linearly: 8 GPUs cost $1.65 per hour on Value.
Is there a cheaper GPU than the Tesla V100 with the same VRAM?
Yes. The NVIDIA RTX A4000 has 16 GB and starts at $0.184 per hour. It differs in architecture (Ampere), memory bandwidth and multi-GPU interconnect, so compare throughput for your workload before switching.
Deploy a Tesla V100 in minutes
Price the exact pod, storage and template in the configurator, add credit, and connect over SSH or JupyterLab. Billed per minute, no commitment.