RTX 3080 vs RTX 5090 x2
Best verified output tok/s per model (single-stream generation). Hover a value for the quant.
RTX 30809 verified runs · best 277.8 tok/s
RTX 5090 x21 verified run · best 195.6 tok/s
| Model | RTX 3080 | RTX 5090 x2 |
|---|
| LFM2.5-8B-A1B | 277.8 Q6_K | — |
| Qwen3.8-27B | 44.4 IQ2_XXS | 195.6 NVFP4 |
| Ornith-1.5-9B | 82.8 Q6_K | — |
| Qwen3.8-27B-Uncensored | 74.1 IQ2_XXS | — |
| Bonsai-27B | 72.9 Q1_0 | — |
| gemma-4-12B-it-qat | 71.2 Q4_0 | — |
| Ornith-1.5-35B-A3B | 64.4 Q4_K_M | — |
| gpt-oss-20b | 57.1 MXFP4 | — |
| Muse-Glimmer-30B | 6.0 Q4_K_M | — |
Head-to-head (models tested on both): RTX 3080 0 · RTX 5090 x2 1 · tied 0 · data as of Sep 23, 2026. Full rules: methodology.
Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board