Wan2.1-T2V-14B
Wan-AI/Wan2.1-T2V-14B
Alibaba Wan 2.1 text-to-video, 720P generation, 14B. Commercial priority model.
Parameters
13.8B
Context
32K
Downloads
680K
Likes
5.1K
Architecture
wan
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 56.4 GB RAM 84.6 · Disk 79.1 | 62.4 GB RAM 93.6 · Disk 105.1 | 72.3 GB RAM 108.5 · Disk 144.1 |
| FP88bit | 31.0 GB RAM 46.4 · Disk 59.1 | 35.8 GB RAM 53.7 · Disk 85.1 | 44.6 GB RAM 66.8 · Disk 124.1 |
| INT88bit | 31.0 GB RAM 46.4 · Disk 59.1 | 35.8 GB RAM 53.7 · Disk 85.1 | 44.6 GB RAM 66.8 · Disk 124.1 |
| AWQ4bit | 18.7 GB RAM 32.0 · Disk 49.4 | 23.0 GB RAM 34.5 · Disk 75.4 | 31.2 GB RAM 46.8 · Disk 114.4 |
| GPTQ4bit | 18.9 GB RAM 32.0 · Disk 49.5 | 23.2 GB RAM 34.8 · Disk 75.5 | 31.4 GB RAM 47.1 · Disk 114.5 |
| GGUF4bit | 19.2 GB RAM 32.0 · Disk 49.8 | 23.5 GB RAM 35.3 · Disk 75.8 | 31.7 GB RAM 47.6 · Disk 114.8 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 60.4 video-sec/GPU-h
Effective production throughput ≈ 36.2 video-sec/GPU-h
score 0.6891
1× RTX 3090
24 GB VRAM
≈ 25.5 video-sec/GPU-h
Effective production throughput ≈ 15.3 video-sec/GPU-h
score 0.6418
8× H100 SXM
640 GB VRAM
≈ 46.8 video-sec/GPU-h
Effective production throughput ≈ 28.1 video-sec/GPU-h
score 0.2415
1× RTX 4090
24 GB VRAM
≈ 75 video-sec/GPU-h
Effective production throughput ≈ 45 video-sec/GPU-h
score 0.6418
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 3090
24 GB VRAM
$76.5
/mo · vast
RTX 4090
24 GB VRAM
$99.0
/mo · vast
RTX 4090D
24 GB VRAM
$117
/mo · vast
L4
24 GB VRAM
$147
/mo · vast
A10
24 GB VRAM
$176
/mo · vast