Can the RTX 4090 24 GB run SCAIL-2 (character animation)?
Yes, but only just: FP8 leaves 1.0 GB free on the RTX 4090 24 GB.
FP8 (17.7 GB) fits with only 1.0 GB to spare (about 23.0 GB of 24 GB).
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | wan2.1_14B_SCAIL_2_fp8_scaled.safetensors FP8 · Comfy-Org/SCAIL-2 | models/diffusion_models | 17.7 GB | Download → |
| Text encoder | umt5_xxl_fp8_e4m3fn_scaled.safetensors UMT5-XXL FP8 (scaled) · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/text_encoders | 6.7 GB | Download → |
| VAE | Wan2_1_VAE_bf16.safetensors Wan 2.1 VAE BF16 (Kijai) · Kijai/WanVideo_comfy | models/vae | 0.3 GB | Download → |
| Also needed | clip_vision_h.safetensors CLIP Vision H · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/clip_vision | 1.3 GB | Download → |
| Total download · keep about the same free on disk | 25.9 GB | |||
Alternatives: UMT5-XXL FP16 (11.4 GB). Only for some workflows: SAM 3.1 multiplex FP16 (segmentation) (1.7 GB, character-replacement workflow (masking)). Sizes from Hugging Face, 2026-10-01.
02Every SCAIL-2 file on 24 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 32.8 GB | 38.1 GB | 14.1 GB too big | the original weights | Hugging Face → |
| FP8 ← SAFETENSORS · Comfy-Org | 17.7 GB | 23.0 GB | fits · 1.0 GB spare | practically identical to the original | Hugging Face → |
| INT8 SAFETENSORS · Comfy-Org | 16.7 GB | 22.0 GB | fits · 2.0 GB spare | practically identical to the original | Hugging Face → |
| Q8_0 GGUF · vantagewithai | 18.1 GB | 23.4 GB | fits · 0.6 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · vantagewithai | 14.2 GB | 19.5 GB | fits · 4.5 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · vantagewithai | 13.0 GB | 18.3 GB | fits · 5.7 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · vantagewithai | 11.5 GB | 16.8 GB | fits · 7.2 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · vantagewithai | 9.1 GB | 14.4 GB | fits · 9.6 GB spare | noticeable loss of detail | Hugging Face → |
| Q2_K GGUF · vantagewithai | 7.3 GB | 12.6 GB | fits · 11.4 GB spare | heavy loss; a last resort | Hugging Face → |
“Needed” adds 4.5 GB of working memory for SCAIL-2 and a 0.8 GB reserve. Sizes 2026-10-01.
03Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run SCAIL-2 on a RTX 4090 24 GB, send your time per image and peak VRAM and it will appear here, credited.
04About the RTX 4090 24 GB
The RTX 4090 24 GB is an Ada Lovelace GPU with hardware FP8, so ComfyUI can compute Comfy-Org's FP8 files natively: small and fast. (Plain FP8 files use FP8 maths with the --fast fp8_matrix_mult option.)
24 GB GDDR6X · 384-bit · 1008 GB/s · Ada Lovelace. Everything that runs on it →
05Questions
How much VRAM does SCAIL-2 need?
Around 12.6 GB with the smallest file (Q2_K) and 23.0 GB with an 8-bit file (FP8). The full 16-bit file needs about 38.1 GB.
Which SCAIL-2 file should I download for the RTX 4090 24 GB?
FP8 (17.7 GB) from Comfy-Org/SCAIL-2.
FP8 or Q8_0 for SCAIL-2 on the RTX 4090 24 GB?
FP8 (17.7 GB): the RTX 4090 24 GB computes it natively, while Q8_0 (18.1 GB) is unpacked on the fly.
How much do I need to download for SCAIL-2 on the RTX 4090 24 GB?
25.9 GB: wan2.1_14B_SCAIL_2_fp8_scaled.safetensors, umt5_xxl_fp8_e4m3fn_scaled.safetensors, Wan2_1_VAE_bf16.safetensors, clip_vision_h.safetensors.