Expired50% confidenceFactExact time
All API calls in ml-intern go through Hugging Face Inference Providers, allowing users with an HF Token to access thousands of open-source models via a unified API, with model switching requiring only a single parameter (e.g., --model moonshot/kimi-k2 or --model openai/gpt-5).
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/8/2026
First Seen
Valid until: 10/6/2026(expired)
Sources
HuggingFace Open-Sources ml-intern: An AI Engineer That Reads Papers and Trains Models Autonomously
bilibiliValleyAI6/30/2026
Related Claims
UnverifiedHugging Face's Hub service provides an Inference API, Spaces application hosting, and AutoTrain training infrastructure, and uses a Git-LFS architecture to manage large model files68% similarUnverifiedHugging Face Model Hub上的模型支持PyTorch和TensorFlow两大框架，并提供统一的API接口68% similarUnverifiedOpenRouter.ai采用统一API接口接入数十个主流模型，按实际用量透明计费67% similarPartially VerifiedThe large model service providers supported by One API include OpenAI, Azure OpenAI, Anthropic Claude, Google Gemini, DeepSeek, ByteDance Doubao, Baidu ERNIE Bot, Alibaba Tongyi Qianwen, iFlytek Spark, Zhipu ChatGLM, 360 Zhinao, Tencent Hunyuan, and others.67% similarUnverifiedDevelopers use the model through the Responses API, and OpenAI itself builds the Codex application using the same API and model.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/290045API
curl https://kongchang.com/api/v1/knowledge/claims/290045MCP
get_claim(id=290045)