Can the RTX 3070 8 GB run Wan 2.1 T2V 14B?
Not entirely. Even Q4_K_M is 6.4 GB over 8 GB, so ComfyUI keeps part of it in system RAM.
The smallest sensible file, Q4_K_M (10.1 GB), needs about 14.4 GB - 6.4 GB more than RTX 3070 8 GB has. ComfyUI can still run it by streaming part of the model from system RAM. How much slower that is depends on your ComfyUI version, the file format and the PCIe link - see the note below.
Also try the FP8 file (14.3 GB) on this GPU. It does not fit entirely, but a current ComfyUI streams the overflow from RAM and native 8-bit files often beat a small GGUF that does fit. Dynamic VRAM →
Using Kijai's WanVideoWrapper instead of the native nodes? Its WanVideo BlockSwap node keeps part of the model in system RAM. To run the FP8 file (14.3 GB) on RTX 3070 8 GB, start with blocks_to_swap ≈ 31 (of 40). Each block is about 0.4 GB; raise the number if you still run out of memory, lower it for speed. estimate WanVideoWrapper →
01With more VRAM
With 20 GB you could run Wan 2.1 14B at 8-bit or better (FP8, 14.3 GB) with no offloading.
What changes from the RTX 3070 8 GB to the RTX 5060 Ti 16 GB → · What changes from the RTX 3070 8 GB to the RTX 3090 24 GB →
02Every Wan 2.1 14B file on 8 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · Comfy-Org | 28.6 GB | 32.9 GB | 24.9 GB too big | the original weights | Hugging Face → |
| Q8_0 GGUF · city96 | 15.9 GB | 20.2 GB | 12.2 GB too big | practically identical to the original | Hugging Face → |
| FP8 SAFETENSORS · Comfy-Org | 14.3 GB | 18.6 GB | 10.6 GB too big | practically identical to the original | Hugging Face → |
| Q6_K GGUF · city96 | 12.5 GB | 16.8 GB | 8.8 GB too big | very close to the original | Hugging Face → |
| Q5_K_M GGUF · city96 | 11.3 GB | 15.6 GB | 7.6 GB too big | close; small differences in fine detail | Hugging Face → |
| Q4_K_M ← GGUF · city96 | 10.1 GB | 14.4 GB | 6.4 GB too big | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · city96 | 7.6 GB | 11.9 GB | 3.9 GB too big | noticeable loss of detail | Hugging Face → |
“Needed” adds 3.5 GB of working memory for Wan 2.1 14B and a 0.8 GB reserve. Sizes 2026-10-01.
03What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | wan2.1-t2v-14b-Q4_K_M.gguf Q4_K_M · city96/Wan2.1-T2V-14B-gguf | models/unet | 10.1 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 | wan_2.1_vae.safetensors Wan 2.1 VAE · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 17.1 GB | |||
Alternatives: UMT5-XXL FP16 (11.4 GB). Sizes from Hugging Face, 2026-10-01.
04Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run Wan 2.1 14B on a RTX 3070 8 GB, send your time per image and peak VRAM and it will appear here, credited.
05About the RTX 3070 8 GB
The RTX 3070 8 GB (Ampere) has no FP8 compute. FP8 files still load and save memory, but ComfyUI converts them back before the maths, so there is no speed gain - Q8_0 GGUF is the closer-to-original 8-bit choice. ComfyUI can compute INT8 files natively on NVIDIA, which is worth a try where a model offers one.
8 GB GDDR6 · 256-bit · 448 GB/s · Ampere. Everything that runs on it →
06Questions
How much VRAM does Wan 2.1 14B need?
Around 11.9 GB with the smallest file (Q3_K_M) and 18.6 GB with an 8-bit file (FP8). The full 16-bit file needs about 32.9 GB.
Which Wan 2.1 14B file should I download for the RTX 3070 8 GB?
Q4_K_M (10.1 GB) from city96/Wan2.1-T2V-14B-gguf. Load it with Unet Loader (GGUF) from the ComfyUI-GGUF node pack.
FP8 or Q8_0 for Wan 2.1 14B on the RTX 3070 8 GB?
Q8_0 (15.9 GB). The RTX 3070 8 GB has no FP8 maths, so FP8 (14.3 GB) only saves memory and is a step further from the original.
How much do I need to download for Wan 2.1 14B on the RTX 3070 8 GB?
17.1 GB: wan2.1-t2v-14b-Q4_K_M.gguf, umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors.