In a standard Claude API response, where does the actual text content of Claude's response appear?
Select an answer to reveal the explanation.
![Claude API response JSON structure infographic. Full response object shown with labeled fields: id, model, stopreason, usage (at top level). Content array highlighted: content[0] = {type: text, text: 'actual response here'}. Path to text shown: response.content[0].text. Contrast with OpenAI format. 1024x768.](/media/anthropic/ccao-f/CCAO-F-e1-q120.webp)
Short Explanation
Claude's response text lives in response.content[0].text. You'll find a content array of blocks, and for a simple text reply the first block is type='text'.
Full Explanation
The Anthropic Messages API response structure: the response object contains a 'content' array. Each element in the content array is a content block with a 'type' field. For text responses, content[0] is {'type': 'text', 'text': 'the actual response text'}. The response also includes 'id', 'model', 'stop_reason', 'stop_sequence', and 'usage' at the top level. This differs from OpenAI's structure (choices[0].message.content). Understanding this structure is fundamental for parsing Claude's responses. The option claiming a top-level 'text' field is wrong. The option using response.choices[0].message.content follows OpenAI's API structure, not Anthropic's. The option using response.output.answer is invented.