diff --git a/docs.json b/docs.json index 53b7e698..ce91aa37 100644 --- a/docs.json +++ b/docs.json @@ -415,6 +415,7 @@ "runpodctl/reference/runpodctl-serverless", "runpodctl/reference/runpodctl-template", "runpodctl/reference/runpodctl-hub", + "runpodctl/reference/runpodctl-model", "runpodctl/reference/runpodctl-network-volume", "runpodctl/reference/runpodctl-registry", "runpodctl/reference/runpodctl-gpu", diff --git a/runpodctl/overview.mdx b/runpodctl/overview.mdx index 8fa4d651..6b10eb84 100644 --- a/runpodctl/overview.mdx +++ b/runpodctl/overview.mdx @@ -182,6 +182,7 @@ Runpod CLI organizes commands into groups based on the resource type: | `runpodctl serverless` | `sls` | Manage Serverless endpoints | | `runpodctl template` | `tpl` | List, search, and manage templates | | `runpodctl hub` | | Browse and deploy from the Runpod Hub | +| `runpodctl model` | | List available models | | `runpodctl network-volume` | `nv` | Manage network volumes | | `runpodctl registry` | `reg` | Manage container registry authentications | | `runpodctl gpu` | | List available GPUs | diff --git a/runpodctl/reference/runpodctl-model.mdx b/runpodctl/reference/runpodctl-model.mdx new file mode 100644 index 00000000..0e34b9fc --- /dev/null +++ b/runpodctl/reference/runpodctl-model.mdx @@ -0,0 +1,143 @@ +--- +title: "model" +sidebarTitle: "model" +--- + +Manage models in the Model Repo, including adding, uploading, listing, and removing models. + +```bash +runpodctl model [flags] +``` + +## Subcommands + +### Add a model + +Register a new model with the Model Repo: + +```bash +runpodctl model add --name "my-model" +``` + +To upload model files from a local directory: + +```bash +runpodctl model add --name "my-model" --model-path ./model-files +``` + +#### Add flags + + +Model name. + + + +Model owner namespace (user or team owner ID). Use this to add models to a team account. + + + +Initial model status. + + + +Credential type (if required for model access). + + + +Credential reference (if required for model access). + + + +Directory containing model files to upload. When specified, all files in the directory are uploaded to the Model Repo. + + + +Create an upload session without uploading files. Use this to get upload URLs for manual file uploads. + + + +File name for single-file upload. + + + +File size in bytes for single-file upload. + + + +Multipart upload part size in bytes. + + + +Content type for the upload. + + + +GraphQL request timeout. Defaults to 30 minutes for model upload operations. + + +### List models + +List models in your account: + +```bash +runpodctl model list +``` + +Filter by provider or name: + +```bash +runpodctl model list --provider huggingface +runpodctl model list --name "llama" +``` + +#### List flags + + +Filter by model provider. + + + +Filter by model name. + + +### Remove a model + +Remove a model from the Model Repo: + +```bash +runpodctl model remove +``` + +You can specify the model by ID or name: + +```bash +runpodctl model remove my-model +``` + + + +Removing a model permanently deletes it and all associated versions from the Model Repo. This action cannot be undone. + + + +## Environment variables + + +Override the GraphQL API endpoint URL. Defaults to `https://api.runpod.io/graphql`. + + +## Using model version hashes + +To create a Serverless endpoint that uses a model from the Model Repo, reference the model by its version hash: + +```bash +runpodctl serverless create --template-id "tpl_abc123" --name "my-endpoint" \ + --model-reference "abc123def456" +``` + +Run `runpodctl model list` to find the version hash for your model. + +## Related commands + +- [`runpodctl serverless create`](/runpodctl/reference/runpodctl-serverless) +- [`runpodctl hub`](/runpodctl/reference/runpodctl-hub)