Create Model
Create a model.
Tune API Keys are the preferred way to authenticate with the API. You can create an API Key from your Tune Studio Profile → 'Access Keys' (in sidebar)
The organization ID for request, it override the x-org-id header if set.
The id of the model.
The name of the model.
LLM model id which is supported by tune.Studio or any huggingface model.
tracker id of the for model analytics.
The time the model was created.
The time the model was last updated.
metadata for the model.
- QUANTIZATION_AWQ: Aware Quantization.
- QUANTIZATION_GPTQ: GPT Quantization.
- QUANTIZATION_SQUEEZELLM: SqueezeLLM Quantization.
- QUANTIZATION_FP16: FP16 Quantization.
- MODALITY_TEXT: Text.
- MODALITY_IMAGE: Image.
- MODALITY_AUDIO: Audio.
- MODALITY_VIDEO: Video.
- MODALITY_IMAGE_TEXT: Image and Text.
Used to specify the resource requirement for a (job, model).
CPU in milli-core.
Memory in MiB.
GPU Type.
Number of GPUs.
in seconds. Note this is inclusive of jobs and init containers,
enforced by k8s, regardless of max_retries set.
Not supported for models.
max number of retries for a job.
feature gates for the model (internal use).
number of replicas for the model.
State of the model. This is used to track the model's lifecycle. The state is updated by the model controller.
State of the model.
- UNSPECIFIED: Model in unknown state.
- PROVISIONING: model is being created on the cluster.
- ENV_SETUP: running initilization steps.
- AWAITING_STARTUP: starting model container.
- LIVENESS_CHECK: checking model liveness.
- READY: model is ready.
- FAILED: model is in error state.
- TERMINATED: model is terminated can be restarted.
extra message for the state.
true if model have public endpoint else false.
uri of the model in the form of username/model_name.
Model or Serving (Group) ID.
URL to access the model or serving group.
Hostname of the endpoint.
deprecated.
server type of the model, if empty used from DB.
Authorizations
Tune API Keys are the preferred way to authenticate with the API. You can create an API Key from your Tune Studio Profile → 'Access Keys' (in sidebar)
Body
The id of the model.
The name of the model.
LLM model id which is supported by tune.Studio or any huggingface model.
tracker id of the for model analytics.
The time the model was created.
The time the model was last updated.
metadata for the model.
- QUANTIZATION_AWQ: Aware Quantization.
- QUANTIZATION_GPTQ: GPT Quantization.
- QUANTIZATION_SQUEEZELLM: SqueezeLLM Quantization.
- QUANTIZATION_FP16: FP16 Quantization.
QUANTIZATION_UNSPECIFIED, QUANTIZATION_AWQ, QUANTIZATION_GPTQ, QUANTIZATION_SQUEEZELLM, QUANTIZATION_FP16 - MODALITY_TEXT: Text.
- MODALITY_IMAGE: Image.
- MODALITY_AUDIO: Audio.
- MODALITY_VIDEO: Video.
- MODALITY_IMAGE_TEXT: Image and Text.
MODALITY_UNSPECIFIED, MODALITY_TEXT, MODALITY_IMAGE, MODALITY_AUDIO, MODALITY_VIDEO, MODALITY_IMAGE_TEXT Used to specify the resource requirement for a (job, model).
CPU in milli-core.
Memory in MiB.
GPU Type.
Number of GPUs.
in seconds. Note this is inclusive of jobs and init containers,
enforced by k8s, regardless of max_retries set.
Not supported for models.
max number of retries for a job.
number of replicas for the model.
State of the model. This is used to track the model's lifecycle. The state is updated by the model controller.
State of the model.
- UNSPECIFIED: Model in unknown state.
- PROVISIONING: model is being created on the cluster.
- ENV_SETUP: running initilization steps.
- AWAITING_STARTUP: starting model container.
- LIVENESS_CHECK: checking model liveness.
- READY: model is ready.
- FAILED: model is in error state.
- TERMINATED: model is terminated can be restarted.
UNSPECIFIED, PROVISIONING, ENV_SETUP, AWAITING_STARTUP, LIVENESS_CHECK, READY, FAILED, TERMINATED extra message for the state.
true if model have public endpoint else false.
uri of the model in the form of username/model_name.
server type of the model, if empty used from DB.
Response
The id of the model.
The name of the model.
LLM model id which is supported by tune.Studio or any huggingface model.
tracker id of the for model analytics.
The time the model was created.
The time the model was last updated.
metadata for the model.
- QUANTIZATION_AWQ: Aware Quantization.
- QUANTIZATION_GPTQ: GPT Quantization.
- QUANTIZATION_SQUEEZELLM: SqueezeLLM Quantization.
- QUANTIZATION_FP16: FP16 Quantization.
QUANTIZATION_UNSPECIFIED, QUANTIZATION_AWQ, QUANTIZATION_GPTQ, QUANTIZATION_SQUEEZELLM, QUANTIZATION_FP16 - MODALITY_TEXT: Text.
- MODALITY_IMAGE: Image.
- MODALITY_AUDIO: Audio.
- MODALITY_VIDEO: Video.
- MODALITY_IMAGE_TEXT: Image and Text.
MODALITY_UNSPECIFIED, MODALITY_TEXT, MODALITY_IMAGE, MODALITY_AUDIO, MODALITY_VIDEO, MODALITY_IMAGE_TEXT Used to specify the resource requirement for a (job, model).
CPU in milli-core.
Memory in MiB.
GPU Type.
Number of GPUs.
in seconds. Note this is inclusive of jobs and init containers,
enforced by k8s, regardless of max_retries set.
Not supported for models.
max number of retries for a job.
number of replicas for the model.
State of the model. This is used to track the model's lifecycle. The state is updated by the model controller.
State of the model.
- UNSPECIFIED: Model in unknown state.
- PROVISIONING: model is being created on the cluster.
- ENV_SETUP: running initilization steps.
- AWAITING_STARTUP: starting model container.
- LIVENESS_CHECK: checking model liveness.
- READY: model is ready.
- FAILED: model is in error state.
- TERMINATED: model is terminated can be restarted.
UNSPECIFIED, PROVISIONING, ENV_SETUP, AWAITING_STARTUP, LIVENESS_CHECK, READY, FAILED, TERMINATED extra message for the state.
true if model have public endpoint else false.
uri of the model in the form of username/model_name.
server type of the model, if empty used from DB.
Was this page helpful?