Check my GPU →

Wan 2.1 I2V 14B 720P on 24 GB of VRAM

Yes, but only just: FP8 leaves 0.8 GB free on 24 GB. The same image-to-video model trained for 1280×720. Identical file sizes to the 480P version, but 720p frames need far more working memory.

Video model · 16.4BAny 24 GB GPUData 2026-10-01
Runs well
Yes, just.

FP8 (16.4 GB) fits with only 0.8 GB to spare (about 23.2 GB of 24 GB).

Best fileFP8
File size16.4 GB
VRAM needed~23.2 GB
System RAM32 GB+
Memory map · 24 GB GPU23.2 GB / 24 GB
012 GB24 GB
Weights FP8 · 16.4 GBWorking memory · 6.0 GBReserve · 0.8 GBFree · 0.8 GB

Got an RTX 40 or 50? Swap Q8_0 for FP8 (16.4 GB): same quality class, computed natively there.

01What to download

PartFileFolderSize
Modelwan2.1_i2v_720p_14B_fp8_e4m3fn.safetensors
FP8 · Comfy-Org/Wan_2.1_ComfyUI_repackaged
models/diffusion_models16.4 GBDownload →
Text encoderumt5_xxl_fp8_e4m3fn_scaled.safetensors
UMT5-XXL FP8 (scaled) · Comfy-Org/Wan_2.1_ComfyUI_repackaged
models/text_encoders6.7 GBDownload →
VAEwan_2.1_vae.safetensors
Wan 2.1 VAE · Comfy-Org/Wan_2.1_ComfyUI_repackaged
models/vae0.3 GBDownload →
Also neededclip_vision_h.safetensors
CLIP Vision H · Comfy-Org/Wan_2.1_ComfyUI_repackaged
models/clip_vision1.3 GBDownload →
Total download · keep about the same free on disk24.7 GB

Alternatives: UMT5-XXL FP16 (11.4 GB). Sizes from Hugging Face, 2026-10-01.

02Every file, on 24 GB

FileSizeNeededOn this cardQualityDownload
16-bit
SAFETENSORS · Comfy-Org
32.8 GB39.6 GB15.6 GB too bigthe original weightsHugging Face →
Q8_0
GGUF · city96
18.1 GB24.9 GBspills 0.9 GBpractically identical to the originalHugging Face →
FP8 ←
SAFETENSORS · Comfy-Org
16.4 GB23.2 GBfits · 0.8 GB sparepractically identical to the originalHugging Face →
Q6_K
GGUF · city96
14.2 GB21.0 GBfits · 3.0 GB sparevery close to the originalHugging Face →
Q5_K_M
GGUF · city96
12.7 GB19.5 GBfits · 4.5 GB spareclose; small differences in fine detailHugging Face →
Q4_K_M
GGUF · city96
11.3 GB18.1 GBfits · 5.9 GB sparegood; some loss in fine detail and textHugging Face →
Q3_K_M
GGUF · city96
8.6 GB15.4 GBfits · 8.6 GB sparenoticeable loss of detailHugging Face →

“Needed” adds 6 GB of working memory for Wan 2.1 I2V 720P and a 0.8 GB reserve. Sizes 2026-10-01.