Aya 3B — VRAM Requirements
Autor: Jakub Rusinowski · Ostatnia aktualizacja: 21 lipca 2026
How much GPU VRAM you need to run Aya 3B (Tiny Aya) Aya 3B by Cohere locally, a 3B-parameter model. Figures are quantized weights + KV cache + framework overhead from a memory-bandwidth roofline model.
Aya 3B needs about 3 GB VRAM at Q4_K_M.
VRAM by Quantization
| Quant | Bits/weight | Weights | Total VRAM |
|---|---|---|---|
| Q2_K | 2.63 | 1.0 GB | 1.8 GB |
| Q3_K_M | 3.41 | 1.3 GB | 2.1 GB |
| Q4_K_M | 4.83 | 1.8 GB | 2.6 GB |
| Q5_K_M | 5.67 | 2.1 GB | 2.9 GB |
| Q6_K | 6.56 | 2.5 GB | 3.3 GB |
| Q8_0 | 8.50 | 3.2 GB | 4.0 GB |
| F16 | 16.00 | 6.0 GB | 6.8 GB |
Switch quantization in the interactive calculator, or see the full Aya 3B (Tiny Aya) model page.
Add this badge to your model card
Model creators: paste this into your Hugging Face model card README to link readers straight to this VRAM breakdown.
[](https://llmconfigurator.com/en/vram-calculator/tiny-aya-3b)Estimates only — actual VRAM varies with context length, batch size, runtime and KV-cache settings.