Get Model List

Retrieve all models currently available on the platform, grouped at two levels: model capability and model provider. Once you obtain a modelId, you can reference it in other APIs that require a specific model.

This API is rate-limited to 60 requests/minute per account. The platform model list changes infrequently, so caching the result on the caller side is recommended.

Request Method

GET

Request URL

https://api-${endpoint}.gptbots.ai/v1/model/list

Request Authentication

This API uses account-level DevKey / DevSecret credentials, passed via HTTP Basic Authentication:

                      
                      Authorization: Basic base64(${DevKey}:${DevSecret})

                    

Join DevKey and DevSecret with a colon, Base64-encode the whole string, and prefix it with Basic (note the space).

To obtain your DevKey / DevSecret: log in to the console and go to Personal Center - Account - Developer Info, where you can find the Developer ID (DevKey) and API DevSecret.

⚠️ Do not confuse this with Agent-level authentication: APIs such as the conversation APIs use an Agent's API Key (Authorization: Bearer ${API Key}), while this API uses the account's DevKey / DevSecret with Basic authentication. The two are not interchangeable.

Base64 is reversible encoding, not encryption. Treat the encoded string like a plaintext password — do not commit it to code repositories or include it in public documents.

Request

Request Example

                      
                      curl -X GET 'https://api-${endpoint}.gptbots.ai/v1/model/list' \
-H 'Authorization: Basic ${base64(DevKey:DevSecret)}'

                    

Request Headers

Request Parameters

None.

Response

Response Example

                      
                      {
  "code": 0,
  "message": "OK",
  "data": {
    "CHAT": {
      "OPEN_AI": [
        { "aiModelVersion": "gpt-4o", "modelId": "65f2c1a8d3b4e5f601234567" },
        { "aiModelVersion": "gpt-3.5-turbo", "modelId": "65f2c1a8d3b4e5f601234568" }
      ],
      "ANTHROPIC_CLAUDE": [
        { "aiModelVersion": "claude-opus-4-8", "modelId": "65f2c1a8d3b4e5f601234569" }
      ]
    },
    "EMBEDDING": {
      "OPEN_AI_EMB": [
        { "aiModelVersion": "text-embedding-3-large", "modelId": "65f2c1a8d3b4e5f60123456a" }
      ]
    },
    "RERANK": { },
    "SPEECH2TEXT": { },
    "TEXT2SPEECH": { },
    "MODERATION": { },
    "ANONYMIZATION": { }
  }
}

                    

Success Response

Two-level structure of data

                      
                      data
 └── model capability      outer key, 7 kinds, fixed order
      └── model provider    inner key, e.g. OPEN_AI
           └── [ model list ]

                    

The outer keys are model capabilities, fixed to the following 7 kinds and always returned in this order:

The inner keys are model provider identifiers. Taking CHAT as an example, possible values include OPEN_AI, AZURE, ANTHROPIC_CLAUDE, GEMINI, ALI_QWEN, DEEP_SEEK, KIMI, META_LLAMA, MISTRAL, ZHIPU_CHATGLM, HUNYUAN, ERINE, XAI_GROK, BYTEDANCE_SEED, SENSE, etc. Provider identifiers differ under other capabilities (e.g. OPEN_AI_EMB, JINA_EMB under EMBEDDING).

• If a capability currently has no available models, its value is an empty object {}. The key is still present and never omitted, so check for emptiness before iterating.
• The platform may add or remove providers over time. Always rely on the actual API response instead of hardcoding the provider list.
• Providers and models are sorted by the platform's configured display priority. The order is for display only — do not build business logic on it (e.g. "the first one is the default model").

Failure Response

Error Codes