Home / GPU guide / RTX 5090 D V2 24GB
| Size (16:9) | 5s | 8s | 10s | 15s |
|---|---|---|---|---|
| 416 × 256 0.1MP · flicker-prone | 1 min 12 s VRAM 23G | 1 min 15 s VRAM 23G | 1 min 17 s VRAM 23G | ⚠ 1 min 30 s VRAM 23G |
| 864 × 480 0.4MP | ⚠ 1 min 44 s VRAM 23G | ⚠ 2 min 26 s VRAM 23G | ⚠ 3 min 3 s VRAM 23G | ⚠ 5 min 2 s VRAM 23G |
| 960 × 544 0.5MP | ⚠ 2 min 1 s VRAM 23G | ⚠ 3 min 4 s VRAM 23G | ⚠ 3 min 60 s VRAM 23G | ⚠ 7 min 5 s VRAM 23G |
| 1152 × 640 0.7MP | ⚠ 2 min 42 s VRAM 23G | ⚠ 4 min 42 s VRAM 23G | ⚠ 6 min 28 s VRAM 23G | ⚠ 12 min 45 s VRAM 23G |
| 1376 × 768 1.0MP · above native | ⚠ 4 min 6 s VRAM 23G | ⚠ 8 min 4 s VRAM 23G | ⚠ 11 min 47 s VRAM 23G | ⚠ 25 min 47 s VRAM 23G |
| 1920 × 1088 2.0MP · above native | ⚠ 11 min 48 s VRAM 23G | ⚠ 29 min 13 s VRAM 23G | ⚠ 47 min 18 s VRAM 23G | ✕ 2.0 h VRAM 23G |
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-1080p 5-10s; 8-20 steps; 2-10 min per clip
Yes. Estimated with 32GB RAM + fast-disk: 416×256·10s at ~1 min 17 s per clip. Configs that exceed VRAM fall back to weight offloading (slower, but they still render).
About 1 min 17 s per 416×256·10s 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 8-step v1.0), 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 5090 D V2 24GB pre-filled)
Same tier: RTX 5090 32GB · RTX 5090 D 32GB
Related: Hardware guide · Acceleration compared · 8GB VRAM setups