Side-by-side GPU comparison: specs, memory, compute performance, and live cloud pricing.
Verdict
L4 has 8 GB more VRAM, making it better suited for large models and long context windows. For compute-bound workloads like training, L4 delivers 5.7× higher FP16 throughput. At $0.39/hr vs $1.46/hr, L4 is the more cost-efficient choice for inference. L4 supports a broader range of models (566 vs 502 from this catalog), giving more flexibility.
Specifications
| L4 | P100 | |
|---|---|---|
| VRAM | 24 GB | 16 GB |
| VRAM Type | GDDR6 | HBM2 |
| Memory Bandwidth | 0.3 TB/s | 0.7 TB/s |
| FP16 Performance | 121 TFLOPS | 21 TFLOPS |
| Manufacturer | NVIDIA | NVIDIA |
| FP8 Support | Yes | No |
| FP4 Support | No | No |
Price / Performance
Based on cheapest single-GPU on-demand pricing. Lower $/TFLOP = better compute value; lower $/GB = better memory value.
| L4 | P100 | |
|---|---|---|
| $/hr (cheapest) | $0.39✓ best | $1.46 |
| $/TFLOP (compute value) | $0.0032✓ best | $0.0689 |
| $/GB VRAM (memory value) | $0.0163✓ best | $0.0912 |
Cloud Pricing
Cheapest on-demand price per provider (single GPU).
Model Compatibility
Models from the catalog that fit on each GPU, grouped by required precision.
L4 (566 models)
P100 (502 models)
You might also compare…
Pricing data refreshed hourly · Last updated April 11, 2026 · Browse all comparisons