返回模型列表
GPT-5.3-Codex
openaiopenai/gpt-5.3-codex
API 使用指南
Chat 调用
标准 chat completions 调用。GPT-5.3-Codex 擅长开放式编程 agent 对话。
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"openai/gpt-5.3-codex","messages":[{"role":"user","content":"Write a Python script to monitor system CPU and memory usage"}]}'工具调用
GPT-5.3-Codex 支持函数/工具调用,适用于 agent 工作流——模型可自主调用工具、解析结果并决定下一步。
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model":"openai/gpt-5.3-codex",
"messages":[{"role":"user","content":"Check my GitHub repo for open pull requests"}],
"tools":[{
"type":"function",
"function":{
"name":"list_github_prs",
"description":"List open pull requests for a GitHub repository",
"parameters":{
"type":"object",
"properties":{
"repo":{"type":"string","description":"Repository name"},
"owner":{"type":"string","description":"Repository owner"}
},
"required":["repo","owner"]
}
}
}],
"tool_choice":"auto"
}'流式调用
GPT-5.3-Codex 支持流式实时 token 输出——但针对长时间自主编程任务,非流式通常更好,因为模型会产生较大的结构化输出。
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"openai/gpt-5.3-codex","messages":[{"role":"user","content":"Write a Docker Compose file for a Node.js app with PostgreSQL"}],"stream":true}'扩展推理
GPT-5.3-Codex 支持通过 reasoning 参数进行扩展思维链推理。这对复杂多步调试和架构决策尤为有价值。
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"openai/gpt-5.3-codex","messages":[{"role":"user","content":"Debug this memory leak in the following Go code..."}],"reasoning":{"effort":"high"}}'事实档案 — 本页每条断言可在此回溯来源
| 来源 | URL | 采集于 | |
|---|---|---|---|
| 发布日期 | openai.com ↗ | 2026-05-28 | 已核实 |
| 训练数据截止 | — | — | 未知 |
| 许可 | openai.com ↗ | 2026-05-28 | 已核实 |
| 相比 GPT-5.2-Codex 速度提升 | openai.com ↗ | 2026-05-28 | 已核实 |
| 网络安全风险等级 | openai.com ↗ | 2026-05-28 | 已核实 |
| SWE-Bench Pro | openai.com ↗ | 2026-05-28 | 已核实 |
| Terminal-Bench 2.0 | llm-stats.com ↗ | 2026-05-28 | 待核实 |
| OSWorld-Verified | llm-stats.com ↗ | 2026-05-28 | 待核实 |
| SWE-Lancer IC Diamond | llm-stats.com ↗ | 2026-05-28 | 待核实 |
| GDPval | openai.com ↗ | 2026-05-28 | 已核实 |
| OpenAI 发布 GPT-5.3-Codex — 最强的编程 agent 模型 | openai.com ↗ | 2026-05-28 | 已核实 |
| GPT-5.3-Codex 与 GPT-5.2-Codex 有什么区别? | openai.com ↗ | 2026-05-28 | 待核实 |
| GPT-5.3-Codex 支持视觉输入吗? | openai.com ↗ | 2026-05-28 | 待核实 |