DeepSeek-V3.2

Chat completions on DeepSeek-V3.2 through an OpenAI-compatible endpoint: deepseek-reasoner returns its reasoning with the answer, deepseek-chat answers directly.

POST/v3/llm/deepseek-3p2/chat/completionsTry it

• Price0.01 / 0.02 ICper 1k tokens, in / out
• Served28.4Kcalls

• v3.2
• Active

Input

model

deepseek-reasoner

messages

role: user content: สวัสดีครับ คุณช่วยอะไรได้บ้าง?

max_tokens

4096

temperature

0.7

top_p

0.9

OutputPOST /v3/llm/deepseek-3p2/chat/completions

{
  "id": "233c1fa8-e6ea-4b4f-86b5-2bbccb39fd9e",
  "object": "chat.completion",
  "created": 1764635256,
  "model": "deepseek-reasoner",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello there! How can I assist you today?",
        "reasoning_content": "Hmm, the user just greeted me with "Hello!". This is a straightforward opening greeting, so I should respond warmly and invitingly. I can start with a cheerful "Hello there!" to mirror their friendly tone, then offer my assistance to encourage them to share what they need. Keeping it simple and open-ended works best here since they haven't specified a request yet."
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 109,
    "total_tokens": 121,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "completion_tokens_details": {
      "reasoning_tokens": 96
    },
    "prompt_cache_hit_tokens": 0,
    "prompt_cache_miss_tokens": 12
  },
  "system_fingerprint": "fp_eaab8d114b_prod0820_fp8_kvcache"
}

Every field, under Response

Request

curl -X POST 'https://api.iapp.co.th/v3/llm/deepseek-3p2/chat/completions' \
  -H 'apikey: YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{ \
    "model": "deepseek-reasoner", \
    "messages": [ \
      {"role": "user", "content": "สวัสดีครับ คุณช่วยอะไรได้บ้าง?"} \
    ], \
    "max_tokens": 4096, \
    "temperature": 0.7, \
    "top_p": 0.9 \
  }'

Response

200OK. The fields are explained under Response.

{
  "id": "233c1fa8-e6ea-4b4f-86b5-2bbccb39fd9e",
  "object": "chat.completion",
  "created": 1764635256,
  "model": "deepseek-reasoner",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello there! How can I assist you today?",
        "reasoning_content": "Hmm, the user just greeted me with "Hello!". This is a straightforward opening greeting, so I should respond warmly and invitingly. I can start with a cheerful "Hello there!" to mirror their friendly tone, then offer my assistance to encourage them to share what they need. Keeping it simple and open-ended works best here since they haven't specified a request yet."
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 109,
    "total_tokens": 121,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "completion_tokens_details": {
      "reasoning_tokens": 96
    },
    "prompt_cache_hit_tokens": 0,
    "prompt_cache_miss_tokens": 12
  },
  "system_fingerprint": "fp_eaab8d114b_prod0820_fp8_kvcache"
}

Request

JSON, in the OpenAI chat-completions shape, so an OpenAI client works with the base URL changed.

Headers

apikeystringrequired

Your API key, created under API Keys in the control panel.

Body

modelstringrequired

deepseek-reasoner, thinking mode, or deepseek-chat, the faster direct mode.

messagesarrayrequired

The conversation: objects with role, one of system, user, assistant, and content.

max_tokensinteger

The most tokens to generate. Default 4096, up to 128,000.

streamboolean

true for server-sent events, a token at a time. Default false.

temperaturefloat

0 to 2, default 0.7. Near 0 for code and maths, 1.0 for analysis, 1.3 for conversation and translation, 1.5 for creative writing.

top_pfloat

Nucleus sampling, default 0.9.

Price

Both models cost 0.01 IC per 1K input tokens and 0.02 IC per 1K output tokens.

Response

The OpenAI chat-completion object, as in the panel: the answer is choices[0].message.content and usage holds the billed token counts. With deepseek-reasoner, message.reasoning_content carries the reasoning and usage.completion_tokens_details.reasoning_tokens counts it. With stream: true the same arrives as data: lines, each a chunk with choices[0].delta, ending in data: [DONE].

The OpenAI chat-completion object, as in the panel: the answer is choices[0].message.content and usage holds the billed token counts. With deepseek-reasoner, message.reasoning_content carries the reasoning and usage.completion_tokens_details.reasoning_tokens counts it. With stream: true the same arrives as data: lines, each a chunk with choices[0].delta, ending in data: [DONE].

Details

Measured

Measure	Result
Parameters	685 billion
Context	128,000 tokens
Mathematics and informatics olympiads	Gold-medal level on IMO and IOI problems, as published by the model's authors
Licence	MIT; open weights

Limits

• 5 requests per second and 200 per minute per key.
• Reasoning tokens are billed as output; allow for them when you set max_tokens.