openrouter-paid-fallback付费回退模式

Set OpenClaw to use a paid OpenRouter model first and fall back to free models when quota is exhausted or rate-limited. Use when the user wants a paid OpenRouter model such as gpt-5.4-mini as the primary model with automatic free fallback.

作者: admin | 来源: ClawHub

OpenRouter Paid Fallback

Configure

1. Keep the OpenRouter API key present in config.
Set agents.defaults.model.primary to the requested paid model.
Set agents.defaults.model.fallbacks to:

- openrouter/free - nvidia/nemotron-3-nano-30b-a3b:free - qwen/qwen3-next-80b-a3b-instruct:free - stepfun/step-3.5-flash:free - arcee-ai/trinity-mini:free

4. Restart the gateway.
Verify with openclaw status or /status.

Defaults

- Primary: INLINECODE9
Keep the same fallback list unless the user asks otherwise.

Rules

- Edit only the model-routing keys.
Preserve unrelated config.
If restart fails, check gateway status and remove stale config entries before retrying.

OpenRouter 付费回退

配置

1. 在配置中保留 OpenRouter API 密钥。
将 agents.defaults.model.primary 设置为请求的付费模型。
将 agents.defaults.model.fallbacks 设置为：

- openrouter/free - nvidia/nemotron-3-nano-30b-a3b:free - qwen/qwen3-next-80b-a3b-instruct:free - stepfun/step-3.5-flash:free - arcee-ai/trinity-mini:free

4. 重启网关。
使用 openclaw status 或 /status 进行验证。

默认设置

- 主要模型：openrouter/openai/gpt-5.4-mini
除非用户另有要求，否则保持相同的回退列表。

规则

- 仅编辑模型路由相关的键。
保留无关的配置。
如果重启失败，请检查网关状态并在重试前删除过时的配置条目。

openrouter-paid-fallback付费回退模式