Side-by-side GPU comparison: specs, memory, compute performance, and live cloud pricing.
Verdict
A40 has 16 GB more VRAM, making it better suited for large models and long context windows. A40 supports a broader range of models (624 vs 577 from this catalog), giving more flexibility.
Specifications
| A40 | Gaudi HL-205 | |
|---|---|---|
| VRAM | 48 GB | 32 GB |
| VRAM Type | GDDR6 | HBM2 |
| Memory Bandwidth | 0.7 TB/s | 1.0 TB/s |
| FP16 Performance | 150 TFLOPS | — |
| Manufacturer | NVIDIA | Habana |
| FP8 Support | No | 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.
| A40 | Gaudi HL-205 | |
|---|---|---|
| $/hr (cheapest) | $0.44 | — |
| $/TFLOP (compute value) | $0.0029 | — |
| $/GB VRAM (memory value) | $0.0092 | — |
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.
A40 (624 models)
Gaudi HL-205 (577 models)
You might also compare…
Pricing data refreshed hourly · Last updated April 11, 2026 · Browse all comparisons