Supported Models

The following Cohere model families are available in a Standard Vault. For the deployment flow, see Creating a Vault.

Model family	Type
Command	Generative, including reasoning, multimodal, and translation variants
North	Generative and agentic bundles, including North Mini Code
Embed	Embeddings
Rerank	Reranker
Compass	Embed + Rerank + Vision Parser bundle
Cohere Transcribe	Speech recognition (ASR), including Arabic
Customized Cohere models	Varies

Embed and Rerank models are generally available self-serve. Generative models, bundles, speech, and customized models may require a waitlist. For the specific versions you can deploy, check the model dropdown when creating a vault, see the models overview, or contact Cohere.

Supported GPUs

A Standard Vault can be deployed on a range of GPUs. To confirm the options available for your model, capacity, and region, contact Cohere.

Differences for Encrypted Vaults

Model and GPU availability differs for Encrypted Vaults, since confidential computing depends on specific GPU hardware.

For pricing per model and performance tier, see Pricing.