Datasheet for local AI49 models98 GPUsData read 2026-09-25
No adsNo tracking
Check my GPU →

HunyuanVideo (13B, original) on 32 GB of VRAM

Video model · 13BAny 32 GB GPUData 2026-09-25
Runs well
Yes — and comfortably.

Download 16-bit (25.6 GB). With the model's working memory it needs about 29.9 GB, leaving 2.1 GB spare on 32 GB. Quality: the original weights.

Best file16-bit
File size25.6 GB
VRAM needed~29.9 GB
System RAM48 GB+
Memory map · 32 GB GPU29.9 GB / 32 GB
016 GB32 GB
Weights 16-bit · 25.6 GBWorking memory · 3.5 GBReserve · 0.8 GBFree · 2.1 GB
calculated from real file sizes plus working memory. Not yet measured on this setup. How this works.

01What to download

PartFileFolderSize
Modelhunyuan_video_t2v_720p_bf16.safetensors
16-bit · Comfy-Org/HunyuanVideo_repackaged
models/diffusion_models25.6 GBDownload →
Text encoderllava_llama3_fp8_scaled.safetensors
LLaVA-Llama3 FP8 (scaled) · Comfy-Org/HunyuanVideo_repackaged
models/text_encoders9.1 GBDownload →
Text encoderclip_l.safetensors
CLIP-L · Comfy-Org/HunyuanVideo_repackaged
models/text_encoders0.2 GBDownload →
VAEhunyuan_video_vae_bf16.safetensors
HunyuanVideo VAE BF16 · Comfy-Org/HunyuanVideo_repackaged
models/vae0.5 GBDownload →
Total download · keep about the same free on disk35.5 GB

Alternatives: LLaVA-Llama3 FP16 (16.1 GB). The 16-bit text encoder is a little more faithful; the smaller one is chosen here because it loads faster and needs less RAM. Only for some workflows: LLaVA-Llama3 vision (0.6 GB, I2V only). Sizes read from Hugging Face (2026-09-25). “Download” links start the file directly; the file name links are the same files the official ComfyUI workflows use.

System RAM: 48 GB or more. ComfyUI keeps the model file, the text encoder and the VAE in system RAM and moves them to the GPU as needed. With this set of files that is about 35.5 GB, plus roughly 6 GB for Windows, ComfyUI and a browser: 41.5 GB in total. With less RAM it still runs, but Windows starts swapping to disk and loading gets very slow. calculated

02Every file, on 32 GB

FileSizeNeededOn this cardQualityDownload
16-bit ←
SAFETENSORS · Comfy-Org
25.6 GB29.9 GBfits · 2.1 GB sparethe original weightsHugging Face →
Q8_0
GGUF · city96
14.0 GB18.3 GBfits · 13.7 GB sparepractically identical to the originalHugging Face →
FP8
SAFETENSORS · Kijai
13.2 GB17.5 GBfits · 14.5 GB sparepractically identical to the originalHugging Face →
Q6_K
GGUF · city96
11.0 GB15.3 GBfits · 16.7 GB sparevery close to the originalHugging Face →
Q5_K_M
GGUF · city96
9.4 GB13.7 GBfits · 18.3 GB spareclose; small differences in fine detailHugging Face →
Q4_K_M
GGUF · city96
7.9 GB12.2 GBfits · 19.8 GB sparegood; some loss in fine detail and textHugging Face →
Q3_K_M
GGUF · city96
6.2 GB10.5 GBfits · 21.5 GB sparenoticeable loss of detailHugging Face →

Sizes from the Hugging Face file listing, read 2026-09-25. “Needed” = file + 3.5 GB working memory for this model + 0.8 GB kept free for the system.

03The text encoder

LLaVA-LLaMA-3 8B + CLIP-L: 16.1 GB as 16-bit, 9.1 GB as FP8 (plus CLIP-L (0.25 GB)). ComfyUI encodes the prompt first and can push the encoder out of VRAM before sampling, so it does not have to fit together with the model. On 32 GB the FP8 encoder fits on its own, so prompt encoding stays fast.