Menu docs

Docs / API

Chat Completions

POST /v1/chat/completions dalam format OpenAI, dengan streaming, tools, dan JSON mode.

Terakhir diperbarui


Setup Anda

Mengisi YOUR_API_KEY di kode halaman ini. Key hanya ada di tab ini: tidak dikirim dan tidak disimpan.

Endpoint

POST https://api.cent-ai.dev/v1/chat/completions
Authorization: Bearer cai_<secret>
Content-Type: application/json

Endpoint ini mengikuti format OpenAI Chat Completions. Arahkan SDK atau tool OpenAI apa pun ke base URL https://api.cent-ai.dev/v1.

Request

FieldWajibYang dilakukan Cent AI
modelyaMenentukan model dan harganya. Nama yang tidak dikenal dijawab 404 model_not_found
messagesyaDikirim ke model apa adanya
streamtidaktrue mengembalikan server-sent events, lihat Streaming panjang
stream_options.include_usagetidaktrue menambah chunk usage terakhir, dengan usage.cost, ke stream
max_tokens, max_completion_tokenstidakDikirim apa adanya
tools, tool_choice, response_format, reasoning_effort, dan lainnyatidakDikirim apa adanya

Parameter diteruskan ke model tanpa diubah. Parameter yang tidak didukung model kembali sebagai error dari model, dengan status HTTP yang dikembalikannya.

Contoh

curl https://api.cent-ai.dev/v1/chat/completions \
  -H "Authorization: Bearer $CENT_AI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-5.3","messages":[{"role":"user","content":"Halo"}]}'

Respons

{
  "id": "chatcmpl-9f2c1a",
  "object": "chat.completion",
  "created": 1792051200,
  "model": "glm-5.3",
  "choices": [{
    "index": 0,
    "message": {"role": "assistant", "content": "Halo!"},
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 10000,
    "prompt_tokens_details": {"cached_tokens": 9000},
    "completion_tokens": 800,
    "total_tokens": 10800,
    "cost": "0.000363000"
  }
}

prompt_tokens sudah termasuk token cache; cached_tokens adalah bagian yang dibaca dari cache dan ditagih dengan tarif cache read. usage.cost adalah jumlah yang ditagih, lihat Biaya per request.

Streaming

Dengan stream: true, respons berupa text/event-stream. Chunk mengikuti format OpenAI dan stream diakhiri data: [DONE]:

: keepalive

data: {"id":"chatcmpl-9f2c1a","object":"chat.completion.chunk","created":1792051200,"model":"glm-5.3","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}

data: {"id":"chatcmpl-9f2c1a","object":"chat.completion.chunk","created":1792051200,"model":"glm-5.3","choices":[{"index":0,"delta":{"content":"Halo"},"finish_reason":null}]}

data: {"id":"chatcmpl-9f2c1a","object":"chat.completion.chunk","created":1792051200,"model":"glm-5.3","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: {"id":"chatcmpl-9f2c1a","object":"chat.completion.chunk","created":1792051200,"model":"glm-5.3","choices":[],"usage":{"prompt_tokens":10000,"prompt_tokens_details":{"cached_tokens":9000},"completion_tokens":800,"total_tokens":10800,"cost":"0.000363000"}}

data: [DONE]

Chunk usage hanya dikirim bila stream_options.include_usage bernilai true.

Kode status

StatusKapan
200Sukses, stream atau tidak
400Body request tidak valid (invalid_request)
401, 402, 403, 404, 413, 429Lihat Error 4xx
503Model gagal sebelum byte pertama terkirim, lihat Error 5xx

Kontrak


POST/v1/chat/completions

Chat completions (OpenAI format)

Autentikasi · Key API (Bearer atau x-api-key)

Body request

  • max_completion_tokensintegeropsional

  • max_tokensintegeropsional

  • messagesobject[]wajib

  • modelstringwajib

  • streambooleanopsional

  • stream_optionsobjectopsional

  • stream_options.include_usagebooleanopsional

    Sends the final usage chunk, with usage.cost, to the client.

Respons

  • 200Completion, or an SSE stream of chat.completion.chunk frames ending with data: [DONE].
  • 400Error in the OpenAI shape.
  • 401Error in the OpenAI shape.
  • 402Error in the OpenAI shape.
  • 403Error in the OpenAI shape.
  • 404Error in the OpenAI shape.
  • 413Error in the OpenAI shape.
  • 429Error in the OpenAI shape that may be retried after Retry-After.ulangi
  • 500Error in the OpenAI shape.
  • 503Error in the OpenAI shape that may be retried after Retry-After.ulangi

Header respons

Retry-Afterx-ratelimit-limitx-ratelimit-remainingx-ratelimit-resetx-request-id

Dibuat dari kontrak OpenAPI.