Aya Vision 8B
Multilingual 8B vision-language model from Cohere's open research lab, released March 2025.
About
Aya Vision 8B is Cohere for AI's 8-billion-parameter vision-language model, released in March 2025 as the smaller sibling of Aya Vision 32B. It pairs C4AI Command R7B with a SigLIP 2 vision encoder and is trained to understand text and images across 23 languages, covering what Cohere describes as more than half the world's population.
Cohere's internal testing reported it outperforming Qwen2.5-VL 7B, Gemini Flash 1.5 8B, and Llama 3.2 11B Vision on the AyaVisionBench and m-WildVision benchmarks (vendor-developed and vendor-reported). Weights are downloadable on Hugging Face and Kaggle, but under a CC-BY-NC 4.0 license, so research and academic use is free while commercial use is prohibited.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 114; Trending score 78/100 from researcher assessment.