Access our models directly through our API and pay for what you use, or deploy on Amazon SageMaker.
Get free, rate-limited usage for learning and prototyping. Usage is free until you go into production
Power your business with hosted language models.
Price per token
Price / Unit
$0.4 / 1M tokens
$0.8 / 1M tokens
Embeddings perform best when the text to be embedded is less than 512 tokens. You can create up to 96 embeddings per API call.
If you require dedicated model instances, dedicated support channels, or custom deployment options, get in touch with our sales team
"Cohere excels at delivering high quality, low-latency language AI models and really supporting them. Having Cohere's team as an extension of ours lets us go 10x faster."