Multi-GPU x3 vs RTX 3090 x2
Best verified output tok/s per model (single-stream generation). Hover a value for the quant.
Multi-GPU x34 verified runs · best 63.3 tok/s
RTX 3090 x26 verified runs · best 240.8 tok/s
| Model | Multi-GPU x3 | RTX 3090 x2 |
|---|
| Qwen3.6-27B | 60.3 Q6_K_L | 240.8 AutoRound INT4 |
| Qwen3.8-27B | — | 219.8 AutoRound INT4 + fp8 KV |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B | — | 197.0 NVFP4 |
| Qwen3.6-35B-A3B | 63.3 APEX-I-Balanced | — |
| Qwen3.8-Flash-Next-REAP-256-duo | — | 53.5 Unsloth-Dynamic-Q3_K_XL |
| Ling-3.0-flash | 43.4 AD-IQ4_XXS | — |
| Qwen3.8-Flash-Next | — | 40.1 Unsloth-Dynamic-IQ3_XXS |
| Laguna-S-2.1 | 23.9 UD-IQ4_XS | — |
Head-to-head (models tested on both): Multi-GPU x3 0 · RTX 3090 x2 1 · tied 0 · data as of Sep 23, 2026. Full rules: methodology.
Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board