=== messages-thinking-roundtrip ===
--- INPUT  (POST messages, request.json) ---
{
  "model": "zai-org/GLM-5.3",
  "messages": [
    {
      "role": "user",
      "content": "Reply with exactly OK."
    }
  ],
  "max_tokens": 1024
}

--- OUTPUT (HTTP 200, 1.0 s, pass=True) ---
{"id":"msg_ef8c4d91fcc045a28b060128017b5a0c","type":"message","role":"assistant","content":[{"type":"thinking","thinking":"The user has asked me to reply with exactly \"OK\". This is a simple, clear instruction. I should respond with just \"OK\" - nothing more, nothing less.","signature":""},{"type":"text","text":"OK"}],"model":"zai-org/GLM-5.3","stop_reason":"end_turn","stop_sequence":null,"usage":{"input_tokens":17,"output_tokens":38,"cache_read_input_tokens":0}}

=== messages-thinking-roundtrip-roundtrip ===
--- INPUT  (POST messages, request.json) ---
{
  "model": "zai-org/GLM-5.3",
  "messages": [
    {
      "role": "user",
      "content": "Reply with exactly OK."
    },
    {
      "role": "assistant",
      "content": [
        {
          "type": "thinking",
          "thinking": "The user has asked me to reply with exactly \"OK\". This is a simple, clear instruction. I should respond with just \"OK\" - nothing more, nothing less.",
          "signature": ""
        },
        {
          "type": "text",
          "text": "OK"
        }
      ]
    },
    {
      "role": "user",
      "content": "Again reply with exactly OK."
    }
  ],
  "max_tokens": 1024,
  "stream": false
}

--- OUTPUT (HTTP 200, 1.1 s, pass=True) ---
{"id":"msg_39a9949f67b84706bc00c5ec9bda6414","type":"message","role":"assistant","content":[{"type":"thinking","thinking":"The user is asking me to reply with exactly \"OK\" again. This is a simple, benign request. There's nothing harmful about it. I should comply.","signature":""},{"type":"text","text":"OK"}],"model":"zai-org/GLM-5.3","stop_reason":"end_turn","stop_sequence":null,"usage":{"input_tokens":63,"output_tokens":36,"cache_read_input_tokens":0}}