My Models
My Models is the workspace for model assets that belong to your account. It includes uploaded models and models produced by supported training workflows. Use it when Model Gallery does not contain the model artifact you need.
Before You Upload
Prepare a complete model directory. The upload workflow expects model metadata,
configuration such as config.json, and compatible model weights such as
.safetensors or .bin files.
Large model uploads can take several minutes. Keep the page open until the upload has been accepted, then monitor the model status from the list.
Upload a Model
- In the Console, open Models → My Models.
- Select Upload Model.
- Provide the model information requested by the form.
- Select the model files and submit the upload.
- Return to Model Assets and wait for the model to become available.
Do not start a deployment while the asset is still uploading or processing.
Understand Model Assets
The model list shows the model name, source type, status, and associated deployments. Typical actions include:
- Deploy an available model.
- Review how many deployments use the model.
- Delete an unused model when its current status permits deletion.
If the model was created by fine-tuning, its base model can also be shown in the asset row.
Deploy Your Model
When the model status is available:
- Select Deploy in the model row.
- Confirm the selected model and give the deployment a recognizable name.
- Choose an activated cluster and configure service resources.
- Create the deployment and monitor it from Self-Deploy → Deployments.
See Create Deployment for topology and resource guidance.
Verify the Result
The model is ready for the next workflow when it appears in My Models with an available status and its row presents Deploy. Record the model name and source before creating the deployment.
Troubleshooting
The model is not deployable
Wait for the upload or processing state to finish. If the status becomes failed, verify that the expected configuration and weight files were included.
The model cannot be deleted
Check whether a deployment still references it. Stop and remove dependent deployments before deleting the model asset.