| by gpu tier | Q2_K 4.5 GB | Q4_K_M 6.0 GB | Q5_K_M 7.5 GB | Q6_K 9.0 GB | Q8_0 12.0 GB | FP16 24.0 GB |
|---|---|---|---|---|---|---|
| 8 GB tier rtx 3050 · 4060 | GREAT 52 t/s | OK 40 t/s | ||||
| 12 GB tier rtx 3060 · 4070 | GREAT 77 t/s | GREAT 59 t/s | GREAT 48 t/s | GREAT 40 t/s | ||
| 16 GB tier rtx 4080 · 4060 ti 16g | GREAT 115 t/s | GREAT 88 t/s | GREAT 72 t/s | GREAT 60 t/s | GREAT 46 t/s | |
| 24 GB tier rtx 3090 · 4090 | GREAT 186 t/s | GREAT 143 t/s | GREAT 116 t/s | GREAT 97 t/s | GREAT 74 t/s | |
| 32 GB tier rtx 5090 · m3 max | GREAT 344 t/s | GREAT 265 t/s | GREAT 215 t/s | GREAT 181 t/s | GREAT 137 t/s | GREAT 70 t/s |
| 48 GB tier a6000 · m3 max 64 | GREAT 111 t/s | GREAT 85 t/s | GREAT 69 t/s | GREAT 58 t/s | GREAT 44 t/s | GREAT 23 t/s |
| 80 GB tier h100 · m3 ultra 128 | GREAT 377 t/s | GREAT 290 t/s | GREAT 235 t/s | GREAT 198 t/s | GREAT 150 t/s | GREAT 77 t/s |