Side-by-side GPU comparison: specs, memory, compute performance, and live cloud pricing.
Verdict
L40 has 16 GB more VRAM, making it better suited for large models and long context windows.
Specifications
| Gaudi HL-205 | L40 | |
|---|---|---|
| VRAM | 32 GB | 48 GB |
| VRAM Type | HBM2 | GDDR6 |
| Memory Bandwidth | 1.0 TB/s | 0.9 TB/s |
| FP16 Performance | — | 181 TFLOPS |
| Manufacturer | Habana | NVIDIA |
| FP8 Support | No | Yes |
| FP4 Support | No | No |
Price / Performance
Based on cheapest single-GPU on-demand pricing. Lower $/TFLOP = better compute value; lower $/GB = better memory value.
| Gaudi HL-205 | L40 | |
|---|---|---|
| $/hr (cheapest) | — | $0.82 |
| $/TFLOP (compute value) | — | $0.0045 |
| $/GB VRAM (memory value) | — | $0.0171 |
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.
Gaudi HL-205 (1000 models)
L40 (1000 models)
You might also compare…
Pricing data refreshed hourly · Last updated June 1, 2026 · Browse all comparisons