I could not find more related category than this one. And I did not find the contact of HF support.
The issue I get is just 401 error on my API requests to the Inference API, and few hours ago everything worked well.
I checked my account and tokens and everything is good.
I have created a new fine-grained token and tried it and got 401 error
FAIL > unable to complete request ProviderApiError: Failed to perform inference: an HTTP error occurred when requesting the provider.
InferenceClientProviderApiError: Failed to perform inference: an HTTP error occurred when requesting the provider.
at innerStreamingRequest (file:///Users/i/node_modules/.pnpm/@huggingface+inference@4.6.1/node_modules/@huggingface/inference/dist/esm/utils/request.js:123:15)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async chatCompletionStream (file:///Users/i/node_modules/.pnpm/@huggingface+inference@4.6.1/node_modules/@huggingface/inference/dist/esm/tasks/nlp/chatCompletionStream.js:10:5)
at async HuggingFaceDriver.stream (file:///Users/i/src/nan.web/packages/app/src/llm/Chat/drivers/ChatDriver.js:321:21)
at async HuggingFaceDriver.stream (file:///Users/i/src/nan.web/packages/app/src/llm/Chat/drivers/HuggingFace/HuggingFaceDriver.js:282:10)
at async CoderAgent.process (file:///Users/i/src/nan.web/packages/app/src/llm/agents/Agent/Agent.js:135:20)
at async processInput (file:///Users/i/src/nan.web/packages/app/apps/core/App.js:472:29)
at async processChat (file:///Users/i/src/nan.web/packages/app/apps/core/App.js:518:13)
at async file:///Users/i/src/nan.web/packages/app/apps/core/App.js:578:4 {
httpRequest: {
url: ‘``https://huggingface.co/proxy/router.huggingface.co/novita/v3/openai/chat/completions’``,
method: ‘POST’,
headers: {
Authorization: ‘Bearer [redacted]’,
‘Content-Type’: ‘application/json’,
‘User-Agent’: ‘@huggingface/inference/4.6.1 Node.js/22’
},
body: {
messages: [Array],
stream: true,
model: ‘deepseek/deepseek-v3-turbo’
}
},
httpResponse: { requestId: ‘’, status: 401, body: ‘’ }
}