OpenRouter's gpt-oss-120b rate limited

[SakuraSakuraSakuraChan]

What happened?

I added some OpenRouter models, and most of them work well. Only one model, gpt-oss-120b, doesn’t work for me and returns a 429 error (please refer to the screenshot). I know what 429 means, but it’s clearly not the case here, since this was the first attempt within 24 hours. I also tried the same model and API key with another front end, and it worked fine, so this doesn’t seem to be a rate limit issue.
What I've tried:
Add http referer
Disable stream
Enable Responses API
Still 429

Version Information

ghcr.io/danny-avila/librechat-dev latest 9e4f088e3fd3 2 days ago 1.68GB
ghcr.io/danny-avila/librechat-rag-api-dev-lite latest 3d3178a04e1b 3 weeks ago 1.69GB

Steps to Reproduce

1. Add model in yaml
2. Say Hi

What browsers are you seeing the problem on?

Firefox

Relevant log output

2026-01-22T12:21:33.594Z debug: [BaseClient] userMessage
{
  messageId: "c185e39a-e47e-4574-a79f-813c39553f00",
  parentMessageId: "00000000-0000-0000-0000-000000000000",
  conversationId: "36bc3da7-c56a-4967-8c4e-836473efbe67",
  sender: "User",
  text: "你是什么模型",
  isCreatedByUser: true,
  tokenCount: 7,
}
2026-01-22T12:21:33.632Z debug: [saveConvo] api/app/clients/BaseClient.js - saveMessageToDatabase #saveConvo
2026-01-22T12:21:33.800Z debug: [AgentStream] Client subscribed to 36bc3da7-c56a-4967-8c4e-836473efbe67, resume: false
2026-01-22T12:21:33.801Z debug: [InMemoryEventTransport] subscribe 36bc3da7-c56a-4967-8c4e-836473efbe67: listeners=1
2026-01-22T12:21:33.801Z debug: [InMemoryEventTransport] isFirstSubscriber 36bc3da7-c56a-4967-8c4e-836473efbe67: count=1
2026-01-22T12:21:33.801Z debug: [GenerationJobManager] Replaying 1 buffered events for 36bc3da7-c56a-4967-8c4e-836473efbe67
2026-01-22T12:21:33.801Z debug: [GenerationJobManager] First subscriber ready, resolving promise for 36bc3da7-c56a-4967-8c4e-836473efbe67
2026-01-22T12:23:27.442Z error: [api/server/controllers/agents/client.js #sendCompletion] Operation aborted 429 Provider returned error

Troubleshooting URL: https://js.langchain.com... [truncated]
2026-01-22T12:23:27.443Z error: [api/server/controllers/agents/client.js #sendCompletion] Unhandled error type 429 Provider returned error

Troubleshooting URL: https://js.langchain.... [truncated]
2026-01-22T12:23:27.443Z debug: [spendTokens] conversationId: 36bc3da7-c56a-4967-8c4e-836473efbe67 | Context: message | Token usage:
{
  promptTokens: 10,
  completionTokens: 4,
}



{"attemptNumber":7,"code":429,"error":{"code":429,"message":"Provider returned error","metadata":{"is_byok":false,"provider_name":"OpenInference","raw":"openai/gpt-oss-120b:free is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/integrations"}},"headers":{},"lc_error_code":"MODEL_RATE_LIMIT","level":"error","message":"[api/server/controllers/agents/client.js #sendCompletion] Operation aborted 429 Provider returned error\n\nTroubleshooting URL: https://js.langchain.com/docs/troubleshooting/errors/MODEL_RATE_LIMIT/\n","pregelTaskId":"7072741e-f7f7-5f59-8e47-ce10c510642d","requestID":null,"retriesLeft":0,"stack":"Error: 429 Provider returned error\n\nTroubleshooting URL: https://js.langchain.com/docs/troubleshooting/errors/MODEL_RATE_LIMIT/\n\n    at APIError.generate (/app/node_modules/openai/core/error.js:63:20)\n    at CustomOpenAIClient.makeStatusError (/app/node_modules/openai/client.js:158:32)\n    at CustomOpenAIClient.makeRequest (/app/node_modules/openai/client.js:301:30)\n    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)\n    at async /app/node_modules/@langchain/openai/dist/chat_models.cjs:2411:29\n    at async RetryOperation._fn (/app/node_modules/p-retry/index.js:50:12)","status":429,"timestamp":"2026-01-22T12:23:27.440Z"}

Screenshots

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

From the logs you provided:

openai/gpt-oss-120b:free is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/integrations