Chat message builder API

Builds a normalized OpenAI-style chat message array from system/context/turns/user inputs and also emits the raw ChatML (<|im_start|>) serialization plus a token estimate — the deterministic way to assemble a chat request. Answers 'how do I build a chat messages array?', 'what does the ChatML for this conversation look like?'.

Price$0.02per request

MethodPOST

Route/v1/llm/chatml-build

StatusLive

MIME typeapplication/json

Rate limit120/minute

Cache0s public

llmchatmlmessageschatpromptopenaibuildagent

API URLhttps://x402.hexl.dev/v1/llm/chatml-build

Integration docs

Example request

{
  "system": "Be concise.",
  "turns": [
    {
      "role": "user",
      "content": "Hi"
    },
    {
      "role": "assistant",
      "content": "Hello"
    }
  ],
  "userMessage": "Bye"
}

Example response

{
  "messages": [
    {
      "role": "system",
      "content": "Be concise."
    },
    {
      "role": "user",
      "content": "Hi"
    },
    {
      "role": "assistant",
      "content": "Hello"
    },
    {
      "role": "user",
      "content": "Bye"
    }
  ],
  "chatml": "<|im_start|>system\nBe concise.<|im_end|>\n<|im_start|>user\nHi<|im_end|>\n<|im_start|>assistant\nHello<|im_end|>\n<|im_start|>user\nBye<|im_end|>\n<|im_start|>assistant\n",
  "messageCount": 4,
  "tokens": 41
}

Input schema

{
  "type": "object",
  "properties": {
    "system": {
      "type": "string"
    },
    "context": {
      "type": "string"
    },
    "turns": {
      "type": "array",
      "items": {
        "type": "object"
      }
    },
    "userMessage": {
      "type": "string"
    }
  }
}

Output schema

{
  "type": "object",
  "additionalProperties": true
}