HunyuanVideo VRAM requirements
Tencent's open text-to-video model. Its README gives the memory for two settings, both 129 frames long. Tencent's figures run from 45 GB for HunyuanVideo, 544p to 60 GB for HunyuanVideo, 720p. The cheapest single GPU for the first is the 48 GB RTX A6000 at Thunder Compute ($0.35 an hour); for the second, the 80 GB A100 at Thunder Compute ($1.09).
What Tencent says each setting needs
| Model and setting | Stated GPU memory | Cheapest single GPU with that much | Per hour | |
|---|---|---|---|---|
| HunyuanVideo, 720p720×1280, 129 frames | 60 GB | A100, 80 GB, Thunder Compute | $1.09 | Rent → |
| HunyuanVideo, 544p544×960, 129 frames | 45 GB | RTX A6000, 48 GB, Thunder Compute | $0.35 | Rent → |
| Tencent's recommendationFor better generation quality | 80 GB | A100, 80 GB, Thunder Compute | $1.09 | Rent → |
The memory figures are Tencent'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
These are Tencent's minimums, measured as peak GPU memory; Tencent recommends an 80 GB GPU for better generation quality, so the 80 GB row is the one to rent for final renders.
In Tencent's words
- HunyuanVideo, 720p: “The minimum GPU memory required is 60GB for 720px1280px129f and 45G for 544px960px129f.”
- HunyuanVideo, 544p: “The minimum GPU memory required is 60GB for 720px1280px129f and 45G for 544px960px129f.”
- Tencent's recommendation: “We recommend using a GPU with 80GB of memory for better generation quality.”
Source: the HunyuanVideo 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.