Can the RTX 3080 10 GB run Wan 2.1 T2V 1.3B?
Yes. 16-bit fits on the RTX 3080 10 GB with 4.4 GB to spare.
Take 16-bit (2.8 GB). In use it comes to about 5.6 GB, with 4.4 GB left over on 10 GB.
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | wan2.1_t2v_1.3B_fp16.safetensors 16-bit · Comfy-Org/Wan_2.1_ComfyUI_repackaged | models/diffusion_models | 2.8 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 | 9.8 GB | |||
Alternatives: UMT5-XXL FP16 (11.4 GB). Sizes from Hugging Face, 2026-10-01.
02Every Wan 2.1 1.3B file on 10 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit ← SAFETENSORS · Comfy-Org | 2.8 GB | 5.6 GB | fits · 4.4 GB spare | the original weights | Hugging Face → |
| Q8_0 GGUF · samuelchristlie | 1.5 GB | 4.3 GB | fits · 5.7 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · samuelchristlie | 1.2 GB | 4.0 GB | fits · 6.0 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · samuelchristlie | 1.1 GB | 3.9 GB | fits · 6.1 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · samuelchristlie | 1.0 GB | 3.8 GB | fits · 6.2 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · samuelchristlie | 0.7 GB | 3.5 GB | fits · 6.5 GB spare | noticeable loss of detail | Hugging Face → |
“Needed” adds 2 GB of working memory for Wan 2.1 1.3B 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 Wan 2.1 1.3B on a RTX 3080 10 GB, send your time per image and peak VRAM and it will appear here, credited.
04About the RTX 3080 10 GB
The RTX 3080 10 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.
10 GB GDDR6X · 320-bit · 760 GB/s · Ampere. Everything that runs on it →
05Questions
How much VRAM does Wan 2.1 1.3B need?
Around 3.5 GB with the smallest file (Q3_K_M) and 4.3 GB with an 8-bit file (Q8_0). The full 16-bit file needs about 5.6 GB.
Which Wan 2.1 1.3B file should I download for the RTX 3080 10 GB?
16-bit (2.8 GB) from Comfy-Org/Wan_2.1_ComfyUI_repackaged.
How much do I need to download for Wan 2.1 1.3B on the RTX 3080 10 GB?
9.8 GB: wan2.1_t2v_1.3B_fp16.safetensors, umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors.