This topic created in 57 days ago, the information mentioned may be changed or developed.
我在 cloudflare 上使用 work + AI Gateway 代理的第三方 LLM API 提供商，有概率出现这样的请求体，导致模型验证失败：
{
"model": "gpt-5.2",
"messages": [
{
"role": "user",
"content": "Hi"
}
],
"max_tokens": 1,
"stream": false
}
正常请求体应该是这样的：
{
"model": "gpt-5.2",
"messages": [
{
"role": "user",
"content": "Hi"
}
],
"max_tokens": 1024
}
{
"model": "gpt-5.2",
"messages": [
{
"role": "user",
"content": "Hi"
}
],
"max_tokens": 1,
"stream": false
}
正常请求体应该是这样的：
{
"model": "gpt-5.2",
"messages": [
{
"role": "user",
"content": "Hi"
}
],
"max_tokens": 1024
}
Select Language