Platform For AI:LLM

Get started

Model Gallery offers pre-trained LLMs from the open-source community. You can perform model fine-tuning, distillation, compression, evaluation, and deployment without coding.

• Quick start: Deploy, fine-tune, and evaluate Qwen3 models

• Quick start: Deploy, fine-tune, and evaluate the QwQ-32B model

• Quick start: Fine-tune, evaluate, and deploy Qwen2.5 models

• Quick start: Train, evaluate, compress, and deploy Qwen2.5-Coder models

• Quick start: Fine-tune, evaluate, compress, and deploy a DistilQwen2 model

• Quick start: Deploy and fine-tune Llama-3 series models

• Quick start: Deploy and fine-tune Qwen1.5 series models

• Quick start: Deploy and fine-tune a Mixtral-8x7B MoE model

• Quick start: Deploy and fine-tune Llama 2 models

• Quick start: Continued pre-training for LLMs

• Quick start: Data augmentation and model distillation for LLMs

• Best practices for LLM evaluation

Advanced usage

If Model Gallery does not include the model you need, or if its training and deployment capabilities do not meet your requirements, use Data Science Workshop (DSW) and Deep Learning Containers (DLC) for model fine-tuning and training, then deploy your model in Elastic Algorithm Service (EAS).

Stage	Use cases
Training	Fine-tune a Llama3-8B model
Deployment	Quickly deploy LLMs in EAS

PAI-Lingjun intelligent computing service

PAI-Lingjun intelligent computing service is designed for large-scale deep learning scenarios, providing heterogeneous computing resources and an AI engineering platform.

• Best practices for fully managed Qwen on Lingjun

Data processing

Machine Learning Designer integrates advanced algorithms for processing text, video, and image data to improve training data quality. Text processing algorithms support editing, transforming, deduplicating, and filtering low-quality data samples. Video and image processing algorithms support data cleaning, content filtering, metadata extraction, and caption generation. Built-in data processing templates are available and can be extended through secondary development.

• LLM data processing workflow for GitHub code

• Data processing for LLM (web text data from Wikipedia)

• Data processing for LLM (thesis data from arXiv)

• Data processing for LLM (SFT data from Alpaca-CoT)

• LLM data processing - Alpaca-CoT (SFT data) - DLC component

• Data processing for LLM (GitHub code) - DLC component

• Image-text filtering

• Video data filtering and labeling

More solutions

• LLM-based intent recognition solution