Home / GPU guide / RTX 3080 20GB (modded)
| Size (16:9) | 5s | 8s | 10s | 15s |
|---|---|---|---|---|
| 416 × 256 0.1MP · flicker-prone | 1 min 5 s VRAM 17G | 1 min 14 s VRAM 17G | 1 min 19 s VRAM 18G | 1 min 35 s VRAM 18G |
| 864 × 480 0.4MP | 1 min 49 s VRAM 19G | 2 min 31 s VRAM 19G | 3 min 7 s VRAM 19G | 4 min 57 s VRAM 19G |
| 960 × 544 0.5MP | 2 min 6 s VRAM 19G | 3 min 8 s VRAM 19G | 3 min 60 s VRAM 19G | 6 min 45 s VRAM 19G |
| 1152 × 640 0.7MP | 2 min 47 s VRAM 19G | 4 min 38 s VRAM 19G | 6 min 12 s VRAM 19G | 11 min 33 s VRAM 19G |
| 1376 × 768 1.0MP · above native | 4 min 6 s VRAM 19G | 7 min 35 s VRAM 19G | 10 min 45 s VRAM 19G | 22 min 13 s VRAM 19G |
| 1920 × 1088 2.0MP · above native | 10 min 45 s VRAM 19G | 24 min 58 s VRAM 19G | 39 min 24 s VRAM 19G | ✕ 1.6 h VRAM 19G |
Times are per whole clip (model load and text encoding included); the first run is slower due to kernel compilation. ⚠ = weight offloading / noticeably slower; ✕ = not recommended.
--fast-disk768p 5s; 4-8 steps; 1-5 min per clip
Yes. Estimated with 32GB RAM + fast-disk: 1920×1088·10s at ~39 min 24 s per clip. Configs that exceed VRAM fall back to weight offloading (slower, but they still render).
About 39 min 24 s per 1920×1088·10s clip (including model load and encoding). Longer or denser clips scale quadratically in the attention term.
At least 64GB of system RAM recommended (minimum 32GB plus a 16GB pagefile). With 32GB, always enable --fast-disk or weights spill into the pagefile.
Stack step distillation (e.g. LightX2V Turbo 4-step v1.1 768p), enable SageAttention / Kitchen attention; for long clips try SLA / VSA sparse attention. All of it is switchable in step 2 of the calculator.
→ Run the numbers for your RAM and budget in the calculator (RTX 3080 20GB (modded) pre-filled)
Same tier: RTX 3070 Ti 8GB · RTX 3080 10GB · RTX 3080 12GB · RTX 3080 Ti 12GB · RTX 4060 Ti 16GB · RTX 4070 12GB · RTX 4070 Super 12GB · RTX 4070 Ti 12GB
Related: Hardware guide · Acceleration compared · 8GB VRAM setups