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

Can the Arc B580 12 GB run Wan 2.1 VACE 14B?

Video model · 17.3B12 GB GDDR6 · 192-bit · 456 GB/s · BattlemageData 2026-09-25
Offload
Not entirely in VRAM, but it can still run.

Nothing fits entirely, but Q3_K_S (7.8 GB) overflows by only about 0.6 GB. ComfyUI keeps that part in system RAM automatically: slower than a full fit, but usable.

Best fileQ3_K_S
File size7.8 GB
VRAM needed~12.6 GB
System RAM32 GB+
Memory map · Arc B580 12 GB12.6 GB needed · 0.6 GB over 12 GB
06 GB13 GB
Weights Q3_K_S · 7.8 GBWorking memory · 4.0 GBReserve · 0.8 GBSpills to system RAM · 0.6 GB
calculated from real file sizes plus working memory. Not yet measured on this setup. How this works.

01What to download

PartFileFolderSize
ModelWan2.1_14B_VACE-Q3_K_S.gguf
Q3_K_S · QuantStack/Wan2.1_14B_VACE-GGUF
models/unet7.8 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 →
Total download · keep about the same free on disk14.8 GB

Alternatives: UMT5-XXL FP16 (11.4 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. 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: 32 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 14.8 GB, plus roughly 6 GB for Windows, ComfyUI and a browser: 20.8 GB in total. With less RAM it still runs, but Windows starts swapping to disk and loading gets very slow. calculated

02Every Wan VACE 14B file on 12 GB

FileSizeNeededOn this cardQualityDownload
16-bit
SAFETENSORS · Comfy-Org
34.7 GB39.5 GB27.5 GB too bigthe original weightsHugging Face →
Q8_0
GGUF · QuantStack
18.7 GB23.5 GB11.5 GB too bigpractically identical to the originalHugging Face →
Q6_K
GGUF · QuantStack
14.5 GB19.3 GB7.3 GB too bigvery close to the originalHugging Face →
Q5_K_M
GGUF · QuantStack
13.0 GB17.8 GB5.8 GB too bigclose; small differences in fine detailHugging Face →
Q4_K_M
GGUF · QuantStack
11.6 GB16.4 GB4.4 GB too biggood; some loss in fine detail and textHugging Face →
Q3_K_S ←
GGUF · QuantStack
7.8 GB12.6 GBspills 0.6 GBnoticeable loss of detailHugging Face →

Sizes from the Hugging Face file listing, read 2026-09-25. “Needed” = file + 4 GB working memory for this model + 0.8 GB kept free for the system.

03The text encoder

UMT5-XXL: 11.4 GB as 16-bit, 6.7 GB as FP8, 3.7 GB as GGUF Q4_K_M. 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 12 GB the FP8 encoder fits on its own, so prompt encoding stays fast.

04About this GPU

Intel Arc GPUs run ComfyUI through PyTorch's native XPU support on Windows 11 and Linux. Memory works the same as on NVIDIA, so the fit verdicts apply. FP8 files save memory but give no speed-up, and some custom nodes are NVIDIA-only.

Arc B580 12 GB: 12 GB GDDR6 · 192-bit · 456 GB/s · Battlemage. Everything that runs on the Arc B580 12 GB →

05What more VRAM would change

With 24 GB you could run Wan VACE 14B at 8-bit or better (Q8_0, 18.7 GB) with no offloading: for example on the RTX 4090 24 GB, RTX 3090 24 GB, RX 7900 XTX 24 GB.

What changes from the Arc B580 12 GB to the RTX 5060 Ti 16 GB →

06Measured and reported results

Nobody has sent measured numbers for this pair yet. If you run Wan VACE 14B on a Arc B580 12 GB, send your time per image and peak VRAM and it will appear here, credited.

07Questions

How much VRAM does Wan VACE 14B need?

Around 12.6 GB with the smallest file (Q3_K_S) and 23.5 GB with an 8-bit file (Q8_0), counting working memory and a small system reserve. The full 16-bit file needs about 39.5 GB.

Which Wan VACE 14B file should I download for the Arc B580 12 GB?

Q3_K_S (7.8 GB) from QuantStack/Wan2.1_14B_VACE-GGUF. Load it with Unet Loader (GGUF) from the ComfyUI-GGUF node pack.

Is FP8 faster than GGUF on the Arc B580 12 GB?

No. This GPU has no FP8 compute, so ComfyUI converts FP8 weights back before the maths. FP8 only saves memory here; Q8_0 GGUF is the closer-to-original 8-bit pick.

How much do I need to download for Wan VACE 14B on the Arc B580 12 GB?

About 14.8 GB for the model file, text encoder and VAE listed on this page, and the same again free on disk. With a 12 GB card, 32 GB of system RAM or more is recommended.