Home / GPU guide / RTX 3070 Ti 8GB
| Size (16:9) | 5s | 8s | 10s | 15s |
|---|---|---|---|---|
| 416 × 256 0.1MP · flicker-prone | 1 min 50 s VRAM 7G | 2 min 9 s VRAM 7G | 2 min 23 s VRAM 7G | 3 min 1 s VRAM 7G |
| 864 × 480 0.4MP | 3 min 36 s VRAM 7G | 5 min 25 s VRAM 7G | 7 min 3 s VRAM 7G | 12 min 23 s VRAM 7G |
| 960 × 544 0.5MP | 4 min 20 s VRAM 7G | 7 min 7 s VRAM 7G | 9 min 35 s VRAM 7G | 17 min 53 s VRAM 7G |
| 1152 × 640 0.7MP | 6 min 9 s VRAM 7G | 11 min 27 s VRAM 7G | 16 min 12 s VRAM 7G | 33 min 10 s VRAM 7G |
| 1376 × 768 1.0MP · above native | 9 min 52 s VRAM 7G | 20 min 30 s VRAM 7G | 30 min 34 s VRAM 7G | 1.1 h VRAM 7G |
| 1920 × 1088 2.0MP · above native | 30 min 37 s VRAM 7G | 1.3 h VRAM 7G | ✕ 2.1 h VRAM 7G | ✕ 5.5 h VRAM 7G |
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-disk --disable-mmap --lowvram480-540p 5s; 4-8 steps; 3-10 min per clip; start at 736x416
Yes. Estimated with 32GB RAM + fast-disk: 1920×1088·8s at ~1.3 h per clip. Configs that exceed VRAM fall back to weight offloading (slower, but they still render).
About 1.3 h per 1920×1088·8s clip (including model load and encoding). Longer or denser clips scale quadratically in the attention term.
At least 32GB of system RAM recommended (minimum 16GB plus a 48GB 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 3070 Ti 8GB pre-filled)
Same tier: 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 · RTX 5060 Ti 16GB
Related: Hardware guide · Acceleration compared · 8GB VRAM setups