Aya Expanse 8B
by Cohere · aya family
8B
parameters
text-generation multilingual creative-writing summarization
Aya Expanse 8B is Cohere's multilingual model optimized for 23 languages, providing strong performance across diverse linguistic tasks. It excels at translation, summarization, and creative writing in languages often underserved by other models. At 6.5 GB VRAM (Q4) it fits on most consumer GPUs and 8 GB+ Macs. A great choice for multilingual workloads where language coverage matters more than raw benchmark scores.
Quick Start with Ollama
ollama run 8b-q4_K_M | Creator | Cohere |
| Parameters | 8B |
| Architecture | transformer-decoder |
| Context | 8K tokens |
| Released | Oct 15, 2025 |
| License | CC-BY-NC-4.0 |
| Ollama | aya-expanse:8b |
Quantization Options
| Format | File Size | VRAM Required | Quality | Ollama Tag |
|---|---|---|---|---|
| Q4_K_M rec | 5.5 GB | 6.5 GB | | 8b-q4_K_M |
| Q8_0 | 9 GB | 10.5 GB | | 8b-q8_0 |
Compatible Hardware
Q4_K_M requires 6.5 GB VRAM