---
title: "Changes to Create Endpoint"
method: POST
path: "/api/v2/inference/endpoints"
---

# Changes to Create Endpoint

`POST /api/v2/inference/endpoints`

> Every recorded change to this operation, newest first.

## Timeline

Changed in 4 of 28 revisions.

- **2026-08-02** `016014e16810`: 11 info
- **2026-06-08** `cbe3ad94235c`: 7 warning, 27 info
- **2026-04-03** `74dc97653bc3`: 8 info
- **2026-04-01** `526c3dd3d9d7`: 1 info

## Changes

- **2026-08-02** `016014e16810`: 11 info
  - added the new optional request property `execution_mode`
  - added the new optional request property `lora_adapters`
  - added the new optional request property `managed_engine`
  - added the new optional request property `queue_timeout_sec`
  - added the new optional request property `source_ref`
  - added the new optional request property `source_ref_branch`
  - added the new optional request property `source_type`
  - the `idle_timeout_sec` request property default value changed from `300` to `60`
  - the `max_concurrency` request property default value changed from `4` to `1`
  - the `max_workers` request property default value changed from `4` to `3`
  - the `min_workers` request property default value changed from `0` to `1`
- **2026-06-08** `cbe3ad94235c`: 7 warning, 27 info
  - the `max_workers` request property's max was set to `32.00`
  - the `min_workers` request property's max was set to `32.00`
  - the `max_workers` request property's min was set to `1.00`
  - the `min_workers` request property's min was set to `0.00`
  - removed the request property `api_format`
  - removed the request property `max_batch_size`
  - removed the request property `max_concurrent`
  - api operation id `api_inference_create_endpoint_api_v2_inference_endpoints_post` removed and replaced with `api_inference_compat_create_endpoint_api_v2_inference_endpoints_post`
  - api tag `Serverless (compat)` added
  - api tag `Inference v2` removed
  - added the new optional request property `cuda_version`
  - added the new optional request property `env`
  - added the new optional request property `gpu_count`
  - added the new optional request property `gpu_tier`
  - added the new optional request property `health_check_path`
  - added the new optional request property `http_port`
  - added the new optional request property `idle_timeout_sec`
  - added the new optional request property `image_ref`
  - added the new optional request property `max_concurrency`
  - added the new optional request property `max_request_bytes`
  - added the new optional request property `model_ref`
  - added the new optional request property `name`
  - added the new optional request property `request_timeout_sec`
  - added the new optional request property `scaling_policy_type`
  - added the new optional request property `scaling_policy_value`
  - added the new optional request property `startup_command`
  - the request property `model_name` became optional
  - the `model_name` request property default value `` was added
  - the `docker_image` request property default value changed from `xcelsior/vllm:latest` to ``
  - the `health_endpoint` request property default value changed from `/health` to ``
  - the `max_workers` request property default value changed from `3` to `4`
  - the `mode` request property default value changed from `sync` to `preset`
  - the `scaledown_window_sec` request property default value `300` was removed
  - request property `scaledown_window_sec` list-of-types was widened by adding types `null` to media type `application/json`
- **2026-04-03** `74dc97653bc3`: 8 info
  - added the new optional request property `api_format`
  - added the new optional request property `docker_image`
  - added the new optional request property `gpu_type`
  - added the new optional request property `health_endpoint`
  - added the new optional request property `max_batch_size`
  - added the new optional request property `max_concurrent`
  - added the new optional request property `mode`
  - added the new optional request property `region`
- **2026-04-01** `526c3dd3d9d7`: 1 info
  - endpoint added

---

[Operation](https://skmtc.dev/aabiro/apis/xcelsior/docs/api/v2/inference/endpoints/post.md) - [API](https://skmtc.dev/aabiro/apis/xcelsior.md) - [Page](https://skmtc.dev/aabiro/apis/xcelsior/changes/api/v2/inference/endpoints/post)