Qianfan Provider: Access Baidu Models via Unified API

Qianfan is Baidu's MaaS offering: a single OpenAI-compatible API that funnels requests to various models through one endpoint and API key. OpenClaw includes it as the official external plugin @openclaw/qianfan-provider.

Property	Value
Provider	`qianfan`
Auth	`QIANFAN_API_KEY`
API	OpenAI-compatible (`openai-completions`)
Base URL	`https://qianfan.baidubce.com/v2`
Default model	`qianfan/deepseek-v4-pro`

Install plugin

Get the official plugin installed, then restart Gateway:

openclaw plugins install @openclaw/qianfan-provider
openclaw gateway restart

Getting started

Create a Baidu Cloud account

Head to the Qianfan Console to sign up or log in, and confirm that Qianfan API access is turned on for your account.

Generate an API key

Pick an existing application or start a new one, then generate an API key. Keys from Baidu Cloud follow the bce-v3/ALTAK-... format.

Run onboarding

openclaw onboard --auth-choice qianfan-api-key

When running non-interactively, the key is pulled from --qianfan-api-key <key> or QIANFAN_API_KEY. During onboarding, the provider config gets written, the QIANFAN alias is added for the default model, and qianfan/deepseek-v4-pro becomes the default model if none is already set.

Verify the model is available

openclaw models list --provider qianfan

Built-in catalog

Model ref	Input	Context	Max output	Reasoning	Notes
`qianfan/deepseek-v4-pro`	text	1,000,000	393,216	Yes	Current DeepSeek flagship
`qianfan/ernie-5.1`	text	128,000	65,536	No	Latest ERNIE text flagship
`qianfan/ernie-5.0`	text, image	128,000	65,536	Yes	Current multimodal and thinking model
`qianfan/deepseek-v3.2`	text	128,000	32,768	No	Deprecated onboarding compatibility default; replaced by `deepseek-v4-pro`
`qianfan/ernie-5.0-thinking-preview`	text, image	128,000	65,536	Yes	Deprecated alias; replaced by `ernie-5.0`

The catalog is fixed; live model discovery is not available.

Tip

Overriding models.providers.qianfan is only necessary if you need a custom base URL or model metadata.

Config example

This example picks the current DeepSeek flagship explicitly, rather than relying on the onboarding compatibility default.

{
  env: { vars: { QIANFAN_API_KEY: "bce-v3/ALTAK-..." } },
  agents: {
    defaults: {
      model: { primary: "qianfan/deepseek-v4-pro" },
      models: {
        "qianfan/deepseek-v4-pro": { alias: "QIANFAN" },
      },
    },
  },
  models: {
    providers: {
      qianfan: {
        baseUrl: "https://qianfan.baidubce.com/v2",
        api: "openai-completions",
        models: [
          {
            id: "deepseek-v4-pro",
            name: "DeepSeek V4 Pro",
            reasoning: true,
            input: ["text"],
            cost: {
              input: 1.771957,
              output: 3.543915,
              cacheRead: 0.147663,
              cacheWrite: 0,
            },
            contextWindow: 1000000,
            maxTokens: 393216,
          },
        ],
      },
    },
  },
}

Note

Model refs carry the qianfan/ prefix (such as qianfan/deepseek-v4-pro).

Transport and compatibility

Qianfan uses the OpenAI-compatible transport path, not native OpenAI request shaping. Standard OpenAI SDK features function normally, though provider-specific parameters might not be passed through.

Troubleshooting

Verify your API key begins with bce-v3/ALTAK- and that Qianfan API access is enabled in the Baidu Cloud console.
If models don't show up, check that the Qianfan service is activated on your account.
Modify the base URL only when using a custom endpoint or proxy.

Model selection, How to choose providers, model refs, and failover behavior.
Configuration reference, The complete OpenClaw configuration reference.
Agent setup, Setting up agent defaults and model assignments.
Qianfan API docs, Official documentation for the Qianfan API.