146 lines
2.2 KiB
Plaintext
146 lines
2.2 KiB
Plaintext
#价值前沿
|
||
|
||
【公司提供的Claude Code API】
|
||
|
||
Claude Code API:
|
||
|
||
sk-ant-oat01-357ef209755a2161be80cd052ebe3985ce3b8aba15d78d84e9b07910a0966b69（已废弃）
|
||
|
||
sk-ant-oat01-d255ca887a1d5b8fd4fa7f37d59cc34cf81c017602acaf27321e6f361a7709bc
|
||
GACCODE 完整请求体 1. HTTP请求格式
|
||
|
||
POST https://154.26.181.99/claudecode/v1/messages
|
||
|
||
|
||
|
||
2. 请求头（Headers）
|
||
|
||
{
|
||
|
||
"anthropic-version": "2023-06-01",
|
||
|
||
"content-type": "application/json",
|
||
|
||
"x-api-key": "sk-ant-oat01-xxxxxxxx...",
|
||
|
||
"Host": "gaccode.com"
|
||
|
||
}
|
||
|
||
说明：因为是IP直连源站（绕过Cloudflare），需要手动设置 Host: gaccode.com
|
||
|
||
让后端nginx/openresty正确路由。SSL验证需要关闭（verify_ssl: False）。
|
||
|
||
|
||
|
||
3. 请求体（Payload）
|
||
|
||
{
|
||
|
||
"model": "claude-opus-4-5-20250514",
|
||
|
||
"max_tokens": 32000,
|
||
|
||
"temperature": 0.7,
|
||
|
||
"messages": [
|
||
|
||
{
|
||
|
||
"role": "user",
|
||
|
||
"content": "你的提示词内容..."
|
||
|
||
}
|
||
|
||
],
|
||
|
||
"stream": true
|
||
|
||
}
|
||
|
||
|
||
|
||
可选字段：
|
||
|
||
{
|
||
|
||
"system": "系统提示词（可选）",
|
||
|
||
"thinking": {
|
||
|
||
"type": "enabled",
|
||
|
||
"budget_tokens": 10000
|
||
|
||
}
|
||
|
||
}
|
||
|
||
注意：thinking 参数是否被gaccode代理支持尚未验证。根据之前的测试，gaccode代理以文本方式返回thinking内容（包裹在<thinking>标签
|
||
|
||
中），说明代理可能不完整支持thinking协议。
|
||
|
||
|
||
|
||
4. curl 完整示例
|
||
|
||
curl -k -X POST "https://154.26.181.99/claudecode/v1/messages" \
|
||
|
||
-H "anthropic-version: 2023-06-01" \
|
||
|
||
-H "content-type: application/json" \
|
||
|
||
-H "x-api-key: sk-ant-oat01-8625cbcbf200d225488cbfcf952f44bf66fac7089f01dd6c336e1b6699939306" \
|
||
|
||
-H "Host: gaccode.com" \
|
||
|
||
-d '{
|
||
|
||
"model": "claude-opus-4-5-20250514",
|
||
|
||
"max_tokens": 32000,
|
||
|
||
"temperature": 0.7,
|
||
|
||
"messages": [{"role": "user", "content": "Hello"}],
|
||
|
||
"stream": true
|
||
|
||
}'
|
||
|
||
|
||
|
||
响应格式
|
||
|
||
{
|
||
|
||
"id": "msg_xxx",
|
||
|
||
"model": "claude-opus-4-6",
|
||
|
||
"content": [
|
||
|
||
{"type": "thinking", "thinking": "内部推理过程..."},
|
||
|
||
{"type": "text", "text": "最终回答..."}
|
||
|
||
],
|
||
|
||
"stop_reason": "end_turn",
|
||
|
||
"usage": {"input_tokens": 139, "output_tokens": 2269}
|
||
|
||
}
|
||
|
||
|
||
|
||
关键约束
|
||
|
||
1. 必须绕过代理: proxies={"http": None, "https": None} 或 --proxy ""
|
||
|
||
2. 并发上限: 3
|
||
|
||
3. 超时建议: 300-600秒
|
||
|
||
4. thinking.budget_tokens 不能超过 max_tokens |