NVIDIA H100 80GB vs NVIDIA RTX Pro 5000 Blackwell 72GB for local LLMs
6 reports on the NVIDIA H100 80GB and 0 on the NVIDIA RTX Pro 5000 Blackwell 72GB
Which is faster for local LLMs?
Nobody has reported the same model at the same quant on both cards yet, so there's no measured head-to-head. Memory bandwidth, which caps how fast a card can write, is 2.5x higher on the NVIDIA H100 80GB (3,350 GB/s on the NVIDIA H100 80GB, 1,344 on the NVIDIA RTX Pro 5000 Blackwell 72GB). The estimates below compare them on the models people run most.
- Memory
- 80GB
- Memory bandwidth
- 3,350 GB/s
- FP16 compute
- 989 TFLOPS
- Reports
- 6
- Memory
- 72GB
- Memory bandwidth
- 1,344 GB/s
- FP16 compute
- 258 TFLOPS
- Reports
- 0
Estimated by the calculator
The quant the calculator recommends for each card and its estimated speed at a 32,768-token context with 32 GB of system RAM, from memory bandwidth and the model's shape.
| Model | NVIDIA H100 80GB | NVIDIA RTX Pro 5000 Blackwell 72GB |
|---|---|---|
| Qwen3.8 27B | 34.1 t/sFP16 | 13.7 t/sFP16 |
| Qwen3.8 125B · 6B active | 202.0 t/sQ4_K_S | 95.0 t/sQ3_K_M |
| DeepSeek V4 Flash 284B · 13B active | 9.6 t/sQ2_K, experts in RAM | 9.2 t/sQ2_K, experts in RAM |
| Qwen3.6 35B · 3B active | 245.7 t/sQ8_0 | 98.6 t/sQ8_0 |
| Qwen3.6 27B | 34.1 t/sFP16 | 13.7 t/sFP16 |
| Gemma 4 26B · 4B active | 87.8 t/sFP16 | 32.7 t/sFP16 |
Change the context, RAM or model in the calculator: NVIDIA H100 80GB · NVIDIA RTX Pro 5000 Blackwell 72GB
Every report on each card, with its source: NVIDIA H100 80GB · NVIDIA RTX Pro 5000 Blackwell 72GB