H100 80GB vs RTX 3090
For running local LLMs · 16 reports across 5 models
Tokens per second by model
| Model | H100 80GB | RTX 3090 |
|---|---|---|
| Qwen3.6up to 35B | 45.0n=1 | 104.0n=10 |
| Qwen2.5 | — | 70.0n=2 |
| DeepSeek V3 | — | —n=1 |
| Gemma 4up to 31B | 125.3n=1 | — |
| Qwen3.5up to 122B | —n=1 | — |