Balanced cloud GPU for inference and graphics at a moderate price point.
Barriers to entry are low here. 10 cloud providers list the A10 with prices starting at just $0.20/hr per GPU (on-demand), making it accessible for testing without a large commitment.
USA 5 configs 1x-4x
USA 1 config 1x
USA 1 config 1x
France 1 config 1x
USA 12 configs 1x-2x
USA 1 config 1x
USA 1 config 1x
USA 3 configs 1x-4x
France 3 configs 1x-4x
Singapore 26 configs 1x-8x
Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.
24GB GDDR6 with Ampere Tensor Cores. Good balance of price and performance for inference. Wide cloud availability and mature driver support.
PCIe-only, no NVLink. 24GB VRAM limits model sizes. Lacks FP8 support found in Ada Lovelace GPUs (L4, L40S). For new inference deployments, consider the L4.
The median on-demand price across providers has fallen about 5% since August 2025, from $1.50 to $1.42/hr per GPU. This reflects the market-wide median, which moves both when providers change prices and when lower or higher priced offerings enter the market.
With 24GB of VRAM, the A10 can typically run 7B to 13B models in FP16, or larger models in 4-bit quantized form.
The A10 has 24GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
The A10 has 600 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
The A10 supports 6 precision formats. Training: BF16, FP16, TF32, FP32. Inference: INT4, INT8.
No. The A10 is a PCIe-only GPU with no NVLink, so it is better suited to single-GPU inference and smaller-scale workloads than large distributed training jobs.
A10 pricing currently ranges from $0.18/hr to $4.52/hr per GPU, depending on the provider, instance type, and billing model.
At 720 hours per month, one A10 can cost between $126.58 to $3,254.40 per month, depending on the provider. Reserved and spot pricing can lower that further.
The A10 is available from 10 cloud providers, including OVHcloud, Cerebrium, Runcrate. Pricing and availability vary by region and billing model.
Yes. We currently track 54 A10 listings across 10 cloud providers:
| Billing type | Listings | Avg $/GPU/hr |
|---|---|---|
| On-demand | 31 | $1.84/hr |
| Reserved | 20 | $1.63/hr |
| Spot | 3 | $0.69/hr |
| FP32 | 31.2 TF |
| TF32 Tensor Core | 62.5 TF / 125 TF |
| BFLOAT16 Tensor Core | 125 TF / 250 TF |
| FP16 Tensor Core | 125 TF / 250 TF |
| INT8 Tensor Core | 250 TOPS / 500 TOPS |
| INT4 Tensor Core | 500 TOPS / 1000 TOPS |
| RT Cores | 72 |
| Encode / Decode | 1 encoder, 1 decoder (+AV1 decode) |
| GPU Memory | 24 GB GDDR6 |
| GPU Memory Bandwidth | 600 GB/s |
| Interconnect | PCIe Gen4: 64 GB/s |
| Form Factor | 1-slot FHFL |
| Max TDP Power | 150W |
| vGPU Software Support | NVIDIA vPC/vApps, NVIDIA RTX™ vWS, NVIDIA Virtual Compute Server (vCS) |
| Secure and Measured Boot with Hardware Root of Trust | Yes |
| NEBS Ready | Level 3 |
| Power Connector | PEX 8-pin |
Source: official Nvidia A10 datasheet.
Last updated
