Chat Format Converter

Paste OpenAI messages JSON and convert to Anthropic, Gemini, or plain text. Saves time when porting chats across models.

Read the full guide: The Complete Guide to OpenAI Message Formatting & LLM Data Processing →

Privacy: processed locally, never uploaded.

↓ Paste in the input area below to see results instantly

Paste an OpenAI messages JSON array and convert to Anthropic, Gemini, or plain text.

Target format

OpenAI Messages (JSON)

Converted output

{
  "system": "You are helpful.",
  "messages": [
    {
      "role": "user",
      "content": "Hi"
    }
  ]
}

Notes

Input shape

Root is an array; each item has role (system/user/assistant) and string content. developer role is supported.

Paste OpenAI messages JSON and convert to Anthropic, Gemini, or plain text. Saves time when porting chats across models.

Quick start

Paste JSON

Root array with role and content per item.
Pick target

Anthropic lifts system; Gemini maps roles to user/model.

System messages

Anthropic merges system lines; Gemini writes systemInstruction.

Typical Workflow

When migrating conversations between AI models, first copy the messages array from OpenAI API response. Paste it into the tool, then select the target format (e.g. Claude format for Anthropic). The converted content can be directly used in the new model's API calls, saving manual rewriting of role labels and message structures.

For debugging, first convert OpenAI format to plain text using this tool. This allows quick inspection of conversation flow without expanding JSON layers. For models like Gemini, converted messages automatically include @user/@bot tags to preserve multi-turn dialog integrity.

Examples

Example

Input

[{"role":"user","content":"Hi"}]

Output

Anthropic messages JSON

FAQ

Tool / function messages?

String content only; adjust complex tool payloads manually.

Why do converted Gemini messages become longer?

Gemini requires explicit user/AI role labels, adding prefixes to each message. This is normal and doesn't affect API functionality. For compact output, choose 'Plain Text' format with custom delimiters.