MCP Server Coreshub

🚀 Distributed Training and Inference Service API

This project provides a set of APIs for retrieving distributed training tasks, their detailed logs, inference services, and inference service logs.

✨ Features

1. Get Distributed Training Tasks (get_distributed_training)

• end_time: End time in nanosecond timestamp format (e.g., 1745304819402256896).
• start_time: Start time in nanosecond timestamp format (e.g., 1745283219402259200).
• fuzzy: Whether to perform a fuzzy query.
• reverse: Whether to reverse the sorting.
• size: Number of items per page.

2. Get Distributed Training Task Detailed Logs (get_distributed_training_detail_log)

• end_time: End time in nanosecond timestamp format (e.g., 1745304819402256896).
• start_time: Start time in nanosecond timestamp format (e.g., 1745283219402259200).
• fuzzy: Whether to perform a fuzzy query.
• reverse: Whether to reverse the sorting.
• size: Number of items per page.

3. Get Inference Services (get_inference_service)

• zone: Zone identifier obtained from the context. Optional values are: xb3, xb2, hb2.
• owner: Username.
• key_words: Keywords.
• page: Page number.
• size: Number of items per page.

4. Get Inference Service Logs (get_inference_service_log)

• zone: Zone identifier obtained from the context. Optional values are: xb3, xb2, hb2.
• owner: Username.
• service_id: Service ID.
• size: Number of items per page.
• reverse: Whether to reverse the sorting.
• start_time: Start UTC time (format: %Y-%m-%dT%H:%M:%S.000Z).
• end_time: End UTC time (format: %Y-%m-%dT%H:%M:%S.000Z).

📚 Documentation

Parameter Explanation

Return Value Explanation

• All functions return a list or detailed information of the corresponding resources.