客户端配置

复制模板前，先选对客户端

这里放终端和 SDK 的配置模板。图文操作请看 OpenCode、WorkBuddy、CC Switch 的独立教程。

终端与客户端配置

使用 API 密钥里的配置方式

不用通读本页。找到你正在使用的客户端，复制对应模板，把 sk-your-api-key 换成自己的 Key，再发一句 Reply with exactly: ok 测试。OpenAI 兼容接口默认用 https://1a1api.top/v1；如果你的网络访问 .top 不稳定，再尝试备用 https://1a1api.com/v1。模型 ID 推荐 gpt-5.5，GPT-5.5 也能识别。

Codex CLI

写入 ~/.codex/config.toml 和 ~/.codex/auth.json，走 OpenAI Responses。

Codex WebSocket

在 Codex CLI 基础上启用 supports_websockets 与 responses_websockets_v2。

Claude Code（环境变量）

官方推荐用 ANTHROPIC_BASE_URL + ANTHROPIC_AUTH_TOKEN 接入第三方网关，不需要写 ~/.claude/config.toml。

CC Switch

把 1A1API 添加为 Claude / Codex provider，一键切换当前终端工具使用的模型。

OpenCode

写入 opencode.json，模型定义和 provider 一起配置。

OpenClaw

合并到 ~/.openclaw/openclaw.json，不要覆盖已有配置。

Hermes Agent

写入 ~/.hermes/config.yaml 和 ~/.hermes/.env。

~/.codex/config.toml

model_provider = "OpenAI"
model = "gpt-5.5"
review_model = "gpt-5.5"
model_reasoning_effort = "xhigh"
disable_response_storage = true
network_access = "enabled"
windows_wsl_setup_acknowledged = true
model_context_window = 1000000
model_auto_compact_token_limit = 900000

[model_providers.OpenAI]
name = "OpenAI"
base_url = "https://1a1api.top"
wire_api = "responses"
requires_openai_auth = true

~/.codex/auth.json

{
  "OPENAI_API_KEY": "sk-your-api-key"
}

model_reasoning_effort = "xhigh" 是 1A1API / Codex CLI 这类客户端识别的扩展配置项，并非所有官方 OpenAI SDK 的通用参数。直接拿到 OpenAI 官方 SDK 里调用普通模型时，标准取值通常只有 low、medium、high，xhigh 仅在 1A1API 中转 + 兼容客户端的组合下生效。

~/.codex/config.toml

model_provider = "OpenAI"
model = "gpt-5.5"
review_model = "gpt-5.5"
model_reasoning_effort = "xhigh"
disable_response_storage = true
network_access = "enabled"
windows_wsl_setup_acknowledged = true
model_context_window = 1000000
model_auto_compact_token_limit = 900000

[model_providers.OpenAI]
name = "OpenAI"
base_url = "https://1a1api.top"
wire_api = "responses"
supports_websockets = true
requires_openai_auth = true

[features]
responses_websockets_v2 = true

只有客户端版本支持 Responses WebSocket 时再使用这个模板；不确定时优先用普通 Codex CLI 模板。

Claude Code 官方对第三方网关的推荐做法是注入环境变量，而不是手写 ~/.claude/config.toml。下面给出 macOS / Linux 与 Windows 两种写法，任选一种放进 shell 配置即可。

macOS / Linux：~/.zshrc 或 ~/.bashrc

export ANTHROPIC_BASE_URL="https://api.1a1api.top"
export ANTHROPIC_AUTH_TOKEN="sk-your-api-key"
export ANTHROPIC_MODEL="claude-sonnet-4-6"
export ANTHROPIC_DEFAULT_OPUS_MODEL="claude-opus-4-7"
export ANTHROPIC_DEFAULT_SONNET_MODEL="claude-sonnet-4-6"
export ANTHROPIC_DEFAULT_HAIKU_MODEL="claude-haiku-4-5"
export CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1

# 改完后 source ~/.zshrc 或重开终端，再运行
claude

Windows PowerShell（当前会话）

$env:ANTHROPIC_BASE_URL="https://api.1a1api.top"
$env:ANTHROPIC_AUTH_TOKEN="sk-your-api-key"
$env:ANTHROPIC_MODEL="claude-sonnet-4-6"
$env:ANTHROPIC_DEFAULT_OPUS_MODEL="claude-opus-4-7"
$env:ANTHROPIC_DEFAULT_SONNET_MODEL="claude-sonnet-4-6"
$env:ANTHROPIC_DEFAULT_HAIKU_MODEL="claude-haiku-4-5"
$env:CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC="1"

claude

Base URL 用 root，不要追加 /v1/messages。Claude Code 会自己拼端点，所以 ANTHROPIC_BASE_URL 写到 https://api.1a1api.top 即可；写成 https://api.1a1api.top/v1 会出现 /v1/v1/messages 这类 404。

不要再去手写 ~/.claude/config.toml。那是第三方代理或 Codex 风格客户端的写法，官方 Claude Code CLI 并不读取这个文件，照抄会导致看似配置完成但实际还在调官方 Anthropic。如果你用的是 CC Switch 或 Claude Code Router 这类工具，它们会自己生成对应配置，请按它们的文档操作。

OpenCode 示例

{
  "provider": {
    "openai": {
      "options": {
        "baseURL": "https://1a1api.top/v1",
        "apiKey": "sk-your-api-key"
      },
      "models": {
        "gpt-5.5": {
          "name": "GPT-5.5",
          "limit": {
            "context": 1050000,
            "output": 128000
          },
          "options": {
            "store": false
          },
          "variants": {
            "low": {},
            "medium": {},
            "high": {},
            "xhigh": {}
          }
        },
        "gpt-5.4": {
          "name": "GPT-5.4",
          "limit": {
            "context": 1050000,
            "output": 128000
          },
          "options": {
            "store": false
          },
          "variants": {
            "low": {},
            "medium": {},
            "high": {},
            "xhigh": {}
          }
        },
        "gpt-5.4-mini": {
          "name": "GPT-5.4 Mini",
          "limit": {
            "context": 400000,
            "output": 128000
          },
          "options": {
            "store": false
          },
          "variants": {
            "low": {},
            "medium": {},
            "high": {},
            "xhigh": {}
          }
        },
        "gpt-5.3-codex": {
          "name": "GPT-5.3 Codex",
          "limit": {
            "context": 400000,
            "output": 128000
          },
          "options": {
            "store": false
          },
          "variants": {
            "low": {},
            "medium": {},
            "high": {},
            "xhigh": {}
          }
        }
      }
    }
  },
  "agent": {
    "build": {
      "options": {
        "store": false
      }
    },
    "plan": {
      "options": {
        "store": false
      }
    }
  },
  "$schema": "https://opencode.ai/config.json"
}

~/.openclaw/openclaw.json

{
  "models": {
    "mode": "merge",
    "providers": {
      "1a1api-openai": {
        "baseUrl": "https://1a1api.top/v1",
        "apiKey": "sk-your-api-key",
        "api": "openai-completions",
        "authHeader": true,
        "models": [
          {
            "id": "gpt-5.5",
            "name": "GPT-5.5",
            "api": "openai-completions",
            "reasoning": true,
            "input": ["text", "image"],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 272000,
            "maxTokens": 128000
          },
          {
            "id": "gpt-5.4",
            "name": "GPT-5.4",
            "api": "openai-completions",
            "reasoning": true,
            "input": ["text", "image"],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 272000,
            "maxTokens": 128000
          },
          {
            "id": "gpt-5.3-codex",
            "name": "GPT-5.3 Codex",
            "api": "openai-completions",
            "reasoning": true,
            "input": ["text", "image"],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 272000,
            "maxTokens": 128000
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "1a1api-openai/gpt-5.5"
      }
    }
  }
}

如果 openclaw.json 已存在，只合并 models.providers 和 agents.defaults.model 相关片段，不要整文件覆盖。

~/.hermes/config.yaml

model:
  default: "gpt-5.5"
  provider: "custom"
  base_url: "https://1a1api.top/v1"

~/.hermes/.env

OPENAI_API_KEY="sk-your-api-key"

Hermes 官方目前不支持原生 Windows，建议在 macOS、Linux 或 WSL2 中使用。

OpenAI JavaScript SDK

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  baseURL: "https://1a1api.top/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-5.5",
  messages: [
    { role: "user", content: "Reply with exactly: ok" },
  ],
});

console.log(response.choices[0]?.message?.content);

OpenAI Python SDK

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://1a1api.top/v1",
)

response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[
        {"role": "user", "content": "Reply with exactly: ok"}
    ],
)

print(response.choices[0].message.content)