Stable Diffusion 3.5 Medium VRAM requirements
Stability AI's 2.5 billion parameter image model, made to run on consumer GPUs, for images of 0.25 to 2 megapixels. Stability AI states 9.9 GB; the cheapest single GPU with that much is the 16 GB Tesla V100 at Verda (DataCrunch), $0.26 an hour.
What Stability AI says each model needs
| Model and setting | Stated GPU memory | Cheapest single GPU with that much | Per hour | |
|---|---|---|---|---|
| Stable Diffusion 3.5 Medium (2.5B)Full performance, text encoders not counted | 9.9 GB | Tesla V100, 16 GB, Verda (DataCrunch) | $0.26 | Rent → |
The memory figures are Stability 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 9.9 GB leaves out the text encoders, which also need memory unless they run on the CPU, so a card with 12 GB or more leaves room for them. Stability AI states no memory figure for Stable Diffusion 3.5 Large (8.1 billion parameters), so it has no page here.
In Stability AI's words
- Stable Diffusion 3.5 Medium (2.5B): “This model only requires 9.9 GB of VRAM (excluding text encoders) to unlock its full performance”
Source: Stability AI's Stable Diffusion 3.5 announcement, 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.