Written by Jakub Rusinowski · Last updated February 17, 2026
Model library → Aya 3B (Tiny Aya) → Aya 3B
Best 3B multilingual model available. Supports 23+ languages with SOTA quality. Ideal for phones, Raspberry Pi, and offline multilingual assistants. Requires just 2–3 GB VRAM.
Aya 3B needs about 3 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
| Parameters | 3 Billion |
| Context window | 8,192 |
| Architecture | Dense |
| Provider | Cohere |
| Licence | CC-BY-NC 4.0 |
| Specified at | Q4_K_M |
| System RAM | 8 GB |
| Record updated | 2026-02-17 |
CC-BY-NC-4.0 — research / non-commercial only. Research / non-commercial only — this licence does NOT permit shipping a commercial product.
Modelled on a reference NVIDIA RTX 4090 (24 GB), with no KV cache (this record has no published architecture). Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.
| Quant | Weights | VRAM needed | Est. speed | Fit on 24 GB |
|---|---|---|---|---|
| Q2_K | 1.0 GB | 1.8 GB | ~244 tok/s (est.) | Fits comfortably |
| Q3_K_M | 1.3 GB | 2.1 GB | ~223 tok/s (est.) | Fits comfortably |
| Q4_K_M | 1.8 GB | 2.6 GB | ~192 tok/s (est.) | Fits comfortably |
| Q5_K_M | 2.1 GB | 2.9 GB | ~178 tok/s (est.) | Fits comfortably |
| Q6_K | 2.5 GB | 3.3 GB | ~165 tok/s (est.) | Fits comfortably |
| Q8_0 | 3.2 GB | 4.0 GB | ~143 tok/s (est.) | Fits comfortably |
| F16 | 6.0 GB | 6.8 GB | ~93 tok/s (est.) | Fits comfortably |
Want the memory numbers alone, at every quantization level and your own context length? Use the Aya 3B VRAM calculator.
or compare on Vast.ai from $0.35/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
The cheapest catalogued GPU that runs Aya 3B is the Intel Arc B570 (10 GB).
Install Ollama, then run:
ollama run aya:3bWeights on Hugging Face: CohereForAI/aya-3b.
Best for: multilingual, edge devices, phone, translation, chat.
← All Aya 3B (Tiny Aya) models | VRAM calculator | Check your own hardware