教程 · Codex / GPT-5.5
Codex v0.124.0 中使用 GPT-5.5
截至 2026-04-25,Codex v0.124.0 在使用自定义 API(也就是 Taylent 这种中转)时,UI 还选不到 gpt-5.5。本文给出一个临时配置方案,等官方更新后请切回官方方案。
过渡方案 — 本配置会覆盖 Codex 默认的 model catalog。后续 Codex 官方更新内置 GPT-5.5 后,请删除自定义文件改用官方方案。
0. 清掉旧配置(如果你之前装过)
如果按更早版本的教程改过 catalog,先删掉以下两处避免冲突:
- • 删除文件
~/.codex/custom-models.json - • 删除
~/.codex/config.toml中含model_catalog_json的那一行
1. 创建 custom-models.json
在 Codex 配置目录下新建:
文件路径
~/.codex/custom-models.json写入以下内容:
custom-models.json
{ "models": [ { "slug": "gpt-5.5", "display_name": "gpt-5.5", "description": "Custom local catalog entry for GPT-5.5 based on gpt-5.4 metadata.", "default_reasoning_level": "medium", "supported_reasoning_levels": [ { "effort": "low", "description": "Fast responses with lighter reasoning" }, { "effort": "medium", "description": "Balances speed and reasoning depth for everyday tasks" }, { "effort": "high", "description": "Greater reasoning depth for complex problems" }, { "effort": "xhigh", "description": "Extra high reasoning depth for complex problems" } ], "shell_type": "shell_command", "visibility": "list", "supported_in_api": true, "priority": 1, "additional_speed_tiers": ["fast"], "context_window": 272000, "effective_context_window_pct": 0.95, "supports_parallel_tool_calls": true, "image_detail_raw_supported": true } ]}说明:模型 slug = gpt-5.5,上下文窗口 272K,4 档推理 effort(low / medium / high / xhigh),支持并行工具调用。
2. 在 config.toml 中引用
编辑 Codex 主配置文件 ~/.codex/config.toml,加上 model_catalog_json 一行:
~/.codex/config.toml
openai_base_url = "https://api.taylent.com/v1"model = "gpt-5.5"model_reasoning_effort = "high"model_catalog_json = "/Users/<你的用户名>/.codex/custom-models.json"重启 Codex 之后,模型选择面板就会出现 gpt-5.5。
路径分隔符 —
model_catalog_json 的路径要按操作系统写:macOS / Linux 用 /,Windows 用 \\(在 TOML 双引号字符串里转义成两个反斜杠)。# macOS / Linuxmodel_catalog_json = "/Users/toint/.codex/custom-models.json" # Windowsmodel_catalog_json = "C:\\Users\\toint\\.codex\\custom-models.json"环境参考
macOS
26.422.30944 (2080)
codex-cli
0.125.0
配置过程出问题?
邮件 info@taylent.com,工单 5 分钟内响应。