V100 16GB from $0.04 per GPU-hour
The 16 GB Volta card. Fine for fine-tuning small models and for CUDA workloads written before Ampere; look at newer cards if you need quantisation.
Billed per started minute. 64 GB of GPU memory on a 4× node. Persistent storage $0.08 per GB-month. No egress fees.
- GPU memory
- 16GBHBM2
- Bandwidth
- 900GB/sSets token throughput on inference
- Node sizes
- 1× · 2× · 4×Same price per GPU at every size
- Reclaim rate
- <5%Spot instances reclaimed over 30 days
What you get with each V100 16GB node.
| Node | GPU memory | vCPU | RAM | Local NVMe | Spot | On-demand | Action |
|---|---|---|---|---|---|---|---|
| 1× V100 16GB | 16 GB | 6 | 32 GB | 500 GB | $0.04/h | $0.08/h | Configure |
| 2× V100 16GB | 32 GB | 12 | 64 GB | 1 TB | $0.08/h | $0.16/h | Configure |
| 4× V100 16GB | 64 GB | 24 | 128 GB | 2 TB | $0.16/h | $0.32/h | Configure |
GPU-to-GPU link: PCIe Gen3, 32 GB/s. Local NVMe is scratch space wiped when the instance ends; keep anything you need on a persistent disk.
32 V100 16GB GPUs free right now.
Straight from the capacity pool the console books against. Prices are identical in every region — pick the one closest to your data.
A region with zero free GPUs still accepts reserved capacity requests — we hold hardware for a term rather than sell what is already taken.
What fits in 16 GB — and what it costs per hour.
Weights need about 2 GB per billion parameters at FP16 / BF16. We keep 20% of the memory free for activations, the KV cache and the CUDA context, then take the smallest node that still fits.
| Open model | Parameters | FP16 / BF16node · spot price |
|---|---|---|
| Mistral 7BAssistants, RAG, classification | 7.2B | 2× $0.08/h |
| Llama 3.1 8BAssistants, agents, fine-tuning | 8B | 2× $0.08/h |
| Gemma 2 27BHigher-quality assistants | 27B | over 4× |
| Qwen2.5 32BCode and reasoning | 32B | over 4× |
| Mixtral 8x7BMixture of experts, high throughput | 46.7B | over 4× |
| Llama 3.3 70BThe common production baseline | 70B | over 4× |
| Qwen2.5 72BMultilingual, long context | 72B | over 4× |
| Mixtral 8x22BMixture of experts, large capacity | 141B | over 4× |
| Llama 3.1 405BLargest widely used open model | 405B | over 4× |
Real usage depends on context length, batch size and the serving engine: a long context can add tens of gigabytes of KV cache. Treat the table as the floor, not the ceiling — and if a job is close to the limit, take the next node size or quantise one step further.
What a dollar buys on a V100 16GB.
Ranked 10 of 21 data-center models in this catalogue on cost per gigabyte of GPU memory.
GPU memory you get for $1.00 of spot time, per hour.
Bandwidth decides token throughput far more often than raw FLOPS.
Spot price divided by published FP16 tensor throughput.
One GPU on spot, billed per minute, storage excluded.
Against the rest of the market
21 providers · September 2026- Median on-demand elsewhere
- $0.79Our on-demand is 90% below it
- Cheapest on-demand seen
- $0.10Our on-demand $0.08 is 20% below it
- Cheapest spot seen
- $0.06Our spot $0.04 is 33% below it
- 24 hours on one GPU
- $0.96 on spot$1.92 on-demand · $2.40 at the cheapest rate found elsewhere
Index built from published prices for the same GPU across 137 providers (public price index (getdeploying.com)), September 2026. How the guarantee is enforced.
What people run on a V100 16GB.
Quantised inference
Small and medium models in 4-bit or 8-bit form, with the whole model resident in memory.
Computer vision and batch jobs
Detection, classification, embeddings and transcoding at a very low hourly rate.
Anything that can checkpoint
On the spot tier this model is 50% below its own on-demand price, with a 2-minute notice before a reclaim.
NVIDIA V100 16GB, as published by NVIDIA.
Taken from the vendor datasheet. Where a figure is not published, the row is absent rather than estimated.
- Vendor
- NVIDIA
- Architecture
- Volta (2017)
- Segment
- Datacenter
- Memory
- 16 GB HBM2
- Memory bandwidth
- 900 GB/s
- FP16 / BF16 tensor
- 112 TFLOPS
- FP32
- 14 TFLOPS
- CUDA cores
- 5,120
- Board power
- 250 W
- Form factor
- PCIe dual-slot
- Interconnect
- PCIe Gen3, 32 GB/s
- Multi-instance (MIG)
- Not available
- 8-bit float (FP8)
- Not supported — use FP16 or integer quantisation
Source: images.nvidia.com · verified against the vendor document
Software that runs on it
8 templatesPre-built environments, pulled on the node before you land on it. Or bring any OCI image from a public or private registry.
On the spot tier
<5% reclaimed · 30 d- 2 minutes of notice on the metadata endpoint, a webhook and the console
- The instance is stopped, not deleted — disk and IP stay attached
- Auto-relaunch on the next free V100 16GB, or switch the same disk to on-demand
- You save $0.96 a day per GPU against on-demand
If the V100 16GB is not the right fit.
Same memory or more, 25% cheaper on spot
RTX 4000 Ada 20 GB GDDR6 · Ada Lovelace $0.06/h spotNext step up in memory: 20 GB instead of 16 GB
V100 32GB 32 GB HBM2 · Volta $0.06/h spotSame Volta generation
A100 PCIe 80GB 80 GB HBM2e · Ampere $0.09/h spotFaster memory: 1,935 GB/s against 900 GB/s
Or put two of them side by side: V100 16GB vs RTX A4000 · V100 16GB vs RTX 4000 Ada · V100 16GB vs V100
V100 16GB — the questions that come up
How much does a V100 16GB cost per hour?
Spot is $0.04 per GPU-hour and on-demand is $0.08, both billed per minute — every started minute costs the hourly price divided by 60, so $0.0007 on spot. Reserved capacity is $0.06 per GPU-hour on a monthly commitment. Storage and egress are not included in that rate: persistent disks are $0.08 per GB-month and there are no egress fees.
How many GPUs can I put in one instance?
Node sizes are 1×, 2×, 4× — up to 4 V100 16GB GPUs in a single virtual machine, with CPU, RAM and local NVMe scaled with the GPU count. The price per GPU is identical at every node size.
Can it run a 70B model?
Not on a single node of this model: at FP16 a 70B model needs about 140 GB of memory for the weights alone, more than 4× 16 GB leaves free. Quantised to 4 bits it needs about 35 GB — check the memory table above — or pick a model with more memory per GPU.
Which regions have it?
Available now in US East (Virginia), EU Central (Frankfurt). Prices are identical in every region; pick the one closest to your data.
How often is a spot V100 16GB reclaimed?
Over the trailing 30 days, <5% of spot instances on this model were reclaimed. You get a 2-minute notice on the metadata endpoint, through a webhook and in the console; the instance is then stopped, never deleted, and the persistent disk stays attached.
Anything else about this model? Ask an engineer — the same people run the nodes.
A V100 16GB for $0.04 an hour, running in under 60 seconds.
Pay as you go — no contracts, no minimum commitment. Add credit, launch, stop whenever you want.
Billed per minute from the moment the instance is reachable. Minimum credit $40, no subscription. 32 V100 16GB GPUs available right now.