---
title: "Changes to Create a chat completion"
method: POST
path: "/chat/completions"
---

# Changes to Create a chat completion

`POST /chat/completions`

> Every recorded change to this endpoint, newest first.

## Timeline

Changed in 14 of 56 revisions.

- **2026-07-20** `16a952d15e3c` — 4 info
- **2026-07-18** `b8f52abd9b46` — 1 info
- **2026-07-08** `a52394ec1728` — 1 info
- **2026-07-05** `8826c77313d2` — 1 info
- **2026-06-18** `b3724a6ffd56` — 17 info
- **2026-06-11** `e41281e79469` — 1 info
- **2026-05-13** `f229066c04d7` — 2 breaking, 3 warning, 1 info
- **2026-04-28** `b1a0a6dfc381` — 2 info
- **2026-01-22** `3b3a2b57119b` — 1 info
- **2025-11-20** `7b573cae8a96` — 1 info
- **2025-11-15** `9d6f6dd5542d` — 1 breaking, 1 info
- **2025-08-07** `595fdf820935` — 1 info
- **2025-07-26** `00cbec136497` — 1 info
- **2025-07-25** `70bffe5f3ff6` — 8 breaking, 19 warning, 6 info

## Changes

- **2026-07-20** `16a952d15e3c` — 4 info
  - added the optional property `oneOf[#/components/schemas/CreateChatCompletionStreamResponse]/usage/completion_tokens_details` to the response with the `200` status (media type: text/event-stream)
  - added the optional property `oneOf[#/components/schemas/CreateChatCompletionStreamResponse]/usage/prompt_tokens_details` to the response with the `200` status (media type: text/event-stream)
  - added the optional property `usage/completion_tokens_details` to the response with the `200` status (media type: application/json)
  - added the optional property `usage/prompt_tokens_details` to the response with the `200` status (media type: application/json)
- **2026-07-18** `b8f52abd9b46` — 1 info
  - added the new enum value `llamacpp` to the `query` request parameter `provider`
- **2026-07-08** `a52394ec1728` — 1 info
  - added the new enum value `zai` to the `query` request parameter `provider`
- **2026-07-05** `8826c77313d2` — 1 info
  - added the new enum value `nvidia` to the `query` request parameter `provider`
- **2026-06-18** `b3724a6ffd56` — 17 info
  - added the new optional request property `frequency_penalty`
  - added the new optional request property `logit_bias`
  - added the new optional request property `logprobs`
  - added the new optional request property `max_completion_tokens`
  - added the new optional request property `n`
  - added the new optional request property `parallel_tool_calls`
  - added the new optional request property `presence_penalty`
  - added the new optional request property `reasoning_effort`
  - added the new optional request property `response_format`
  - added the new optional request property `seed`
  - added the new optional request property `stop`
  - added the new optional request property `temperature`
  - added the new optional request property `tool_choice`
  - added the new optional request property `top_logprobs`
  - added the new optional request property `top_p`
  - added the new optional request property `user`
  - request property `max_tokens` deprecated
- **2026-06-11** `e41281e79469` — 1 info
  - added the new enum value `minimax` to the `query` request parameter `provider`
- **2026-05-13** `f229066c04d7` — 2 breaking, 3 warning, 1 info
  - added `#/components/schemas/SSEvent, #/components/schemas/CreateChatCompletionStreamResponse` to the response body `oneOf` list for the response status `200` (media type: text/event-stream)
  - the response's body type/format changed from `object`/`` to ``/`` for status `200` (media type: text/event-stream)
  - removed the optional property `data` from the response with the `200` status (media type: text/event-stream)
  - removed the optional property `event` from the response with the `200` status (media type: text/event-stream)
  - removed the optional property `retry` from the response with the `200` status (media type: text/event-stream)
  - added the optional property `choices/items/logprobs` to the response with the `200` status (media type: application/json)
- **2026-04-28** `b1a0a6dfc381` — 2 info
  - added the new optional request property `messages/items/tool_calls/items/extra_content`
  - added the optional property `choices/items/message/tool_calls/items/extra_content` to the response with the `200` status
- **2026-01-22** `3b3a2b57119b` — 1 info
  - added the new enum value `moonshot` to the `query` request parameter `provider`
- **2025-11-20** `7b573cae8a96` — 1 info
  - added the new enum value `ollama_cloud` to the `query` request parameter `provider`
- **2025-11-15** `9d6f6dd5542d` — 1 breaking, 1 info
  - response property `choices/items/message/content` list-of-types was widened by adding types `array` to media type `application/json` of response `200`
  - request property `messages/items/content` list-of-types was widened by adding types `array` to media type `application/json`
- **2025-08-07** `595fdf820935` — 1 info
  - added the new enum value `mistral` to the `query` request parameter `provider`
- **2025-07-26** `00cbec136497` — 1 info
  - added the new enum value `google` to the `query` request parameter `provider`
- **2025-07-25** `70bffe5f3ff6` — 8 breaking, 19 warning, 6 info
  - removed the enum value `google` from the `query` request parameter `provider`
  - removed the enum value `minimax` from the `query` request parameter `provider`
  - removed the enum value `mistral` from the `query` request parameter `provider`
  - removed the enum value `moonshot` from the `query` request parameter `provider`
  - removed the enum value `nvidia` from the `query` request parameter `provider`
  - removed the enum value `ollama_cloud` from the `query` request parameter `provider`
  - removed the enum value `zai` from the `query` request parameter `provider`
  - request property `messages/items/content` list-of-types was narrowed by removing types `array` from media type `application/json`
  - removed the request property `frequency_penalty`
  - removed the request property `logit_bias`
  - removed the request property `logprobs`
  - removed the request property `max_completion_tokens`
  - removed the request property `messages/items/tool_calls/items/extra_content`
  - removed the request property `n`
  - removed the request property `parallel_tool_calls`
  - removed the request property `presence_penalty`
  - removed the request property `reasoning_effort`
  - removed the request property `response_format`
  - removed the request property `seed`
  - removed the request property `stop`
  - removed the request property `temperature`
  - removed the request property `tool_choice`
  - removed the request property `top_logprobs`
  - removed the request property `top_p`
  - removed the request property `user`
  - removed the optional property `choices/items/logprobs` from the response with the `200` status (media type: application/json)
  - removed the optional property `choices/items/message/tool_calls/items/extra_content` from the response with the `200` status (media type: application/json)
  - request property `max_tokens` reactivated
  - removed `#/components/schemas/SSEvent, #/components/schemas/CreateChatCompletionStreamResponse` from the response body `oneOf` list for the response status `200` (media type: text/event-stream)
  - added the optional property `data` to the response with the `200` status (media type: text/event-stream)
  - added the optional property `event` to the response with the `200` status (media type: text/event-stream)
  - added the optional property `retry` to the response with the `200` status (media type: text/event-stream)
  - response property `choices/items/message/content` list-of-types was narrowed by removing types `array` from media type `application/json` of response `200`

---

[Operation](https://skmtc.dev/inference-gateway/apis/inference-gateway-api/docs/chat/completions/post.md) · [API](https://skmtc.dev/inference-gateway/apis/inference-gateway-api.md) · [Page](https://skmtc.dev/inference-gateway/apis/inference-gateway-api/changes/chat/completions/post)