Skip to main content

​
Setup

See our Getting Started guide for initial setup.

​
Upload

You can upload models from Hugging Face or S3 and run inference on a dedicated endpoint through Together AI. For more information, see our Dedicated Inference docs 
together models upload
    --model-name [TEXT]
    --model-source [URI]

​
Options

NameArgumentsDescription
--model-namestringThe name to give to your uploaded model [required]
--model-sourcestringThe source uri of the model [required]
--model-typemodel or adapterWhether the model is a full model or an adapter
--hf-tokenstringHugging Face token (if uploading from Hugging Face)
--descriptionstringA description of your model
--base-modelstringThe base model to use for an adapter if setting it to run against a serverless pool. Only used for model_type ‘adapter’.
--lora-modelstringThe lora pool to use for an adapter if setting it to run against, say, a dedicated pool. Only used for model_type ‘adapter’.
--jsonOutput in JSON format

​
List all models

# List Models
$ together models list

​
Options

NameDescription
--typeFilter models by type. dedicated is the only available option
--jsonOutput in JSON format