Wan2.1 VRAM requirements
Open video models from the Wan team: text-to-video at 1.3 and 14 billion parameters, and image-to-video at 14 billion. Wan-AI states 8.19 GB; the cheapest single GPU with that much is the 16 GB Tesla V100 at Verda (DataCrunch), $0.26 an hour.
What Wan-AI says each model needs
| Model and setting | Stated GPU memory | Cheapest single GPU with that much | Per hour | |
|---|---|---|---|---|
| T2V-1.3B480P text-to-video; a 5-second video on an RTX 4090 in about 4 minutes, per the README | 8.19 GB | Tesla V100, 16 GB, Verda (DataCrunch) | $0.26 | Rent → |
The memory figures are Wan-AI's, not our estimate. The price is the lowest on-demand rate for one GPU with at least that much memory, from our latest check; a card that holds a model is not a promise of its speed.
Before you rent
The README gives the 14B models' memory only in an image of its efficiency table, not in its text, so they have no figure here. Its RTX 4090 example adds --offload_model True and --t5_cpu, which move parts of the model to the CPU, to avoid running out of memory.
In Wan-AI's words
- T2V-1.3B: “The T2V-1.3B model requires only 8.19 GB VRAM, making it compatible with almost all consumer-grade GPUs. It can generate a 5-second 480P video on an RTX 4090 in about 4 minutes”
Source: the Wan2.1 README on GitHub, read 17 October 2026.
Text models are sized by a formula instead: LLM VRAM requirements for every open language model, or the LLM VRAM calculator for your own context and batch.
More models sized by their publishers' figures
- Whisper (speech recognition, OpenAI)
- Stable Diffusion 3.5 Medium (image generation, Stability AI)
- Wan2.2 (video generation, Wan-AI)
- HunyuanVideo (video generation, Tencent)
- olmOCR (OCR and PDF conversion, Ai2)
- MinerU (OCR and document parsing, OpenDataLab)
- GTE-Qwen2 (text embeddings, Alibaba)