A40 vs A100
Explore a head to head comparison of specifications, performance, and pricing.
A40
The NVIDIA A40 delivers high-performance computing capabilities for AI, machine learning, and data science applications.
A100
The NVIDIA A100 is a powerful Ampere-based GPU designed for AI training, inference, and high-performance computing workloads.
A40 vs A100: Which Should You Choose?
The A40 offers 48 GB of VRAM — 1.2× the 40 GB on the A100 — making it better suited for large model workloads that require holding more parameters in GPU memory. On FP16 throughput, the A100 delivers 77.97 TFLOPS versus 37.42 TFLOPS on the A40 — 2× faster for mixed-precision training and inference. Memory bandwidth favors the A40 at 0.70 TB/s compared to 0.00 TB/s on the A100, which directly impacts inference latency for memory-bandwidth-bound models. On Shadeform, the A40 starts from $1.10/hr versus $1.36/hr for the A100 — 24% more expensive — reflecting the performance premium. The A100 is available across 5 cloud providers on Shadeform compared to 2 for the A40, giving more options for region and pricing flexibility.
A40 — Best Use Cases
- •General-purpose deep learning training
- •Fine-tuning models up to 13B parameters
- •AI inference at moderate throughput
- •Computer vision and NLP workloads
Choose A40 when:
- ✓You need 48 GB+ VRAM for large models or long context windows
- ✓Cost efficiency is your primary concern
- ✓Your workload does not require peak FP16 throughput
- ✓Your preferred provider already has availability
A100 — Best Use Cases
- •General-purpose deep learning training
- •Fine-tuning models up to 13B parameters
- •AI inference at moderate throughput
- •Computer vision and NLP workloads
Choose A100 when:
- ✓40 GB VRAM is sufficient for your workload
- ✓Maximum performance justifies the higher cost
- ✓You are training large models or running high-throughput inference
- ✓You need flexibility across multiple cloud providers or regions
See how the A40 & A100 compare
Compare detailed hardware specifications and average pricing for the A40 and A100.
Compare Hardware Specifications
| A40 | A100 | |
|---|---|---|
| GPU Type | A40 | A100 |
| VRAM per GPU | 48 GB | 40 GB |
| Manufacturer | NVIDIA | NVIDIA |
| Architecture | Ampere | Ampere |
| Interconnect | PCIe Gen4 | PCIe Gen4 or SXM4 |
| Memory Bandwidth | 696 GB/s | 1.55 TB/s |
| FP16 TFLOPS | 37.42 TFLOPS (1:1) | 77.97 TFLOPS (4:1) |
| CUDA Cores | 10752 | 6912 |
| Tensor Cores | 336 (3rd Gen) | 432 (3rd Gen) |
| RT Cores | 84 (2nd Gen) | N/A |
| Base Clock | 1125 MHz | 765 MHz |
| Boost Clock | 1740 MHz | 1410 MHz |
| TDP | 300W | 250W-400W |
| Process Node | TSMC 8nm | TSMC 7nm |
| Data Formats | INT8, BF16, FP16, TF32, FP32 | INT8, BF16, FP16, TF32, FP32, FP64 |
Compare Average On-Demand Pricing
| A40 | A100 | |
|---|---|---|
| 1 GPU | $1.48 /hr | $1.88 /hr |
| 2 GPUs | $2.20 /hr | $4.38 /hr |
| 4 GPUs | $5.92 /hr | $8.64 /hr |
| 8 GPUs | $8.80 /hr | $14.90 /hr |
Frequently Asked Questions: A40 vs A100
The main differences are VRAM (48 GB vs 40 GB), FP16 throughput (37.42 vs 77.97 TFLOPS).
The A100 is generally better for large language model training due to its higher throughput and 40 GB of VRAM, which allows fitting larger models or larger batch sizes in a single pass. For smaller models or fine-tuning tasks where cost matters more, both GPUs can be effective.
On Shadeform, the A40 is available from $1.10/hr. The A100 starts from $1.36/hr. Prices vary by provider, region, and contract length. Reserved commitments can reduce hourly costs significantly compared to on-demand pricing.
The A40 has more VRAM at 48 GB, compared to 40 GB on the A100. Higher VRAM allows you to run larger models without quantization, use longer context windows, and process larger batch sizes — all of which improve throughput and reduce latency for memory-bound workloads.
Based on TFLOPS per dollar, the A100 offers better raw compute value at current Shadeform on-demand rates. However, the best choice depends on your specific workload — if you need the extra VRAM or throughput of the A40, paying the premium may be justified by faster job completion and lower total cost.
The A100 is currently available across 5 cloud providers on Shadeform's network, compared to 2 for the A40. Shadeform lets you deploy either GPU across all available providers from a single platform, so you can always find available capacity without manually checking each cloud.
Mixing different GPU types in a single training cluster is generally not recommended, as it creates performance bottlenecks where faster GPUs wait for slower ones. For best results, use a homogeneous cluster of either A40 or A100. Shadeform supports on-demand clusters of up to 64 GPUs of the same type with no commitment required.
Explore A40 & A100 Instances
Browse available instances with A40 and A100 GPUs. Filter by provider, availability, and more to find the perfect instance for your needs.
Explore more GPU comparisons
Select any two GPUs to compare their specifications and explore pricing across providers.
