---
title: "Get Model Endpoint"
method: GET
path: "/v1/llm/model-endpoints/{model_endpoint_name}"
---

# Get Model Endpoint

`GET /v1/llm/model-endpoints/{model_endpoint_name}`

Describe the LLM Model endpoint with given name.

## Path parameters

- `model_endpoint_name` string, required

## Response `200`

Successful Response

- GetLLMModelEndpointV1Response
  - `id` string, required
  - `name` string, required
  - `model_name` string, required
  - `source` 'hugging_face', required
  - `status` 'READY' | 'UPDATE_PENDING' | 'UPDATE_IN_PROGRESS' | 'UPDATE_FAILED' | 'DELETE_IN_PROGRESS', required
  - `inference_framework` 'deepspeed' | 'text_generation_inference' | 'vllm' | 'lightllm' | 'tensorrt_llm' | 'sglang', required
  - `inference_framework_image_tag` string, nullable
  - `num_shards` integer, nullable
  - `quantize` 'bitsandbytes' | 'awq'
  - `checkpoint_path` string, nullable
  - `chat_template_override` string, nullable — A Jinja template to use for this endpoint. If not provided, will use the chat template from the checkpoint
  - `task_expires_seconds` integer, nullable — For async endpoints, how long a task can wait in queue before expiring (in seconds).
  - `queue_message_timeout_seconds` integer, nullable — For async endpoints, the queue message visibility/lock timeout in seconds (SQS VisibilityTimeout / ASB lock_duration). Note: Azure Service Bus has a maximum of 300 seconds; values above this will be clamped.
  - `spec` GetModelEndpointV1Response
    - `id` string, required
    - `name` string, required
    - `endpoint_type` 'async' | 'sync' | 'streaming', required
    - `destination` string, required
    - `deployment_name` string, nullable
    - `metadata` object, nullable
    - `bundle_name` string, required
    - `status` 'READY' | 'UPDATE_PENDING' | 'UPDATE_IN_PROGRESS' | 'UPDATE_FAILED' | 'DELETE_IN_PROGRESS', required
    - `post_inference_hooks` string[], nullable
    - `default_callback_url` string, nullable
    - `default_callback_auth` union
      - CallbackBasicAuth
        - `kind` 'basic', required
        - `username` string, required
        - `password` string, required
      - CallbackmTLSAuth
        - `kind` 'mtls', required
        - `cert` string, required
        - `key` string, required
    - `labels` object, nullable
    - `aws_role` string, nullable
    - `results_s3_bucket` string, nullable
    - `created_by` string, required
    - `created_at` string, date-time, required
    - `last_updated_at` string, date-time, required
    - `deployment_state` ModelEndpointDeploymentState — This is the entity-layer class for the deployment settings related to a Model Endpoint.
      - `min_workers` integer, required
      - `max_workers` integer, required
      - `per_worker` integer, required
      - `concurrent_requests_per_worker` integer, required
      - `forwarder_max_concurrency` integer, nullable
      - `available_workers` integer, nullable
      - `unavailable_workers` integer, nullable
    - `resource_state` ModelEndpointResourceState — This is the entity-layer class for the resource settings per worker of a Model Endpoint. Note: in the multinode case, there are multiple "nodes" per "worker". "Nodes" is analogous to a single k8s pod that may take up all the GPUs on a single machine. "Workers" is the smallest unit that a request can be made to, and consists of one leader "node" and multiple follower "nodes" (named "worker" in the k8s LeaderWorkerSet definition). cpus/gpus/memory/storage are per-node, thus the total consumption by a "worker" is cpus/gpus/etc. multiplied by nodes_per_worker.
      - `cpus` union, required
        - string
        - integer
        - number
      - `gpus` integer, required
      - `memory` union, required
        - string
        - integer
        - number
      - `gpu_type` 'nvidia-tesla-t4' | 'nvidia-ampere-a10' | 'nvidia-ampere-a100' | 'nvidia-ampere-a100e' | 'nvidia-hopper-h100' | 'nvidia-hopper-h100-1g20gb' | 'nvidia-hopper-h100-3g40gb' — Lists allowed GPU types for Launch.
      - `storage` union
        - string
        - integer
        - number
      - `nodes_per_worker` integer, required
      - `optimize_costs` boolean, nullable
    - `num_queued_items` integer, nullable
    - `public_inference` boolean, nullable
    - `task_expires_seconds` integer, nullable
    - `queue_message_timeout_seconds` integer, nullable

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/scaleapi/apis/launch.md) · [All operations](https://skmtc.net/scaleapi/apis/launch/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/scaleapi/launch/versions/8333576dbe43/schema)