Datasetloom: Multimodal LLM Training Data Construction and Evaluation Platform

Datasetloom is an open-source platform for constructing and evaluating datasets for multimodal large language models (VLMs), developed by 599yongyang and updated in December 2025. It provides a full-stack framework using TypeScript, Next.js, and NestJS to streamline the creation of training data for vision-language tasks.

Use Cases

• Constructing multimodal training sets for Vision-Language Models
• Evaluating model performance using the platform's assessment tools
• Managing dataset versions for large language model fine-tuning

Strengths

• MIT licensed for open-source modification and commercial use
• Built on a modern TypeScript stack including Next.js, NestJS, and Shadcn-UI
• Specifically designed for Vision-Language Model (VLM) dataset workflows

Limitations

• Lack of pre-compiled data records in the repository
• Requires a full-stack environment and technical expertise to deploy and use

Provenance

Source

599yongyang (GitHub)

Freshness

Updated as of December 2025.