Aya Expanse — local AI model by Cohere
Written by Jakub Rusinowski · Last updated
The world's best open multilingual model. Aya Expanse supports 23 languages and is specifically optimized for non-English speakers. A breakthrough for global AI accessibility — outperforms Llama 3.1 in Arabic, Chinese, Hindi, French, Spanish, Russian, and 17 more languages.
Variants
The smallest Aya Expanse variant needs about 6 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| Aya Expanse 8B → 8B | ~5.6 GB | 8,192 | ollama run aya-expanse:8b | |
| Aya Expanse 32B → 32B | ~20.3 GB | 128,000 | ollama run aya-expanse:32b |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run Aya Expanse locally
Install Ollama, then pull the tag.
ollama run aya-expanse:8bPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Research / non-commercial only — this licence does NOT permit shipping a commercial product.
Applies to: Aya Expanse 8B, Aya Expanse 32BRecommended GPU
The cheapest catalogued GPU that runs Aya Expanse locally (min 6 GB VRAM) is the Intel Arc B570 (10 GB).
Can I run Aya Expanse on my GPU?
- Aya Expanse on AMD Radeon RX 7900 XTX
- Aya Expanse on AMD Radeon RX 9060 XT 8GB
- Aya Expanse on Apple M1 Pro
- Aya Expanse on Apple M2 Pro
- Aya Expanse on Apple M3 Pro
- Aya Expanse on Apple M4
- Aya Expanse on Apple M5
- Aya Expanse on Intel Arc B570
- Aya Expanse on Intel Arc B580
- Aya Expanse on NVIDIA GeForce RTX 3060 (12GB)
- Aya Expanse on NVIDIA GeForce RTX 3070
- Aya Expanse on NVIDIA GeForce RTX 3070 Ti
- Aya Expanse on NVIDIA GeForce RTX 3080 (10GB)
- Aya Expanse on NVIDIA GeForce RTX 3090
- Aya Expanse on NVIDIA GeForce RTX 4060
- Aya Expanse on NVIDIA GeForce RTX 4070
- Aya Expanse on NVIDIA GeForce RTX 4070 Super
- Aya Expanse on NVIDIA GeForce RTX 4070 Ti
- Aya Expanse on NVIDIA GeForce RTX 4090
- Aya Expanse on NVIDIA GeForce RTX 5060
- Aya Expanse on NVIDIA GeForce RTX 5060 Ti 8GB
- Aya Expanse on NVIDIA GeForce RTX 5070
- Aya Expanse on NVIDIA GeForce RTX 5090
- Aya Expanse on NVIDIA L40S
- Aya Expanse on NVIDIA RTX 6000 Ada Generation
Aya Expanse — frequently asked questions
How much VRAM does Aya Expanse need?
Aya Expanse needs about 6 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Aya Expanse 8B (6 GB, Q4_K_M); Aya Expanse 32B (20 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Aya Expanse on an RTX 4090 (24 GB)?
Yes — Aya Expanse runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Aya Expanse?
Q4_K_M is the best balance of quality and VRAM for Aya Expanse in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Aya Expanse with Ollama?
Install Ollama, then run: ollama run aya-expanse:8b. This downloads Aya Expanse and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.