Radeon AI Pro R9700 x3 vs RTX 5070
Best sourced output tok/s per model (single-stream generation) — engine and version, quantization, context and a source link on every row. Hover a value for the quant.
Radeon AI Pro R9700 x39 sourced runs · best 223.4 tok/s
RTX 50705 verified runs · best 185.8 tok/s
| Model | Radeon AI Pro R9700 x3 | RTX 5070 |
|---|---|---|
| gpt-oss-120b | 223.4 | — |
| NVIDIA-Nemotron-3-Nano-4B | — | 185.8 |
| Apodex-1.1-mini | — | 182.3 |
| Ornith-1.5-9B | — | 102.0 |
| Nemotron-Cascade-2-30B-A3B | 95.4 | — |
| Qwen3.8-Flash-Next | 28.1 | 81.8 |
| Qwen3-VL-8B-Instruct | 58.8 | — |
| Qwen3.8-27B | — | 48.3 |
| Muse-Glimmer-30B | 47.8 | — |
| Ling-3.0-flash | 39.3 | — |
| Ornith-1.5-35B-A3B | 35.4 | — |
| Ornith-1.0-9B | 31.2 | — |
| gemma-4-31B-it | 23.9 | — |