alibaba/x-deeplearning

Health v2 · maintenance only

Adoption not part of the score

4304 stars · 1025 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

X-DeepLearning (XDL) is an industrial deep learning framework from Alibaba optimized for high-dimension sparse data scenarios such as advertising, recommendation, and search. It includes a training engine, the Blaze inference engine, and a deep tree matching (TDM) serving engine.

Use cases

• train deep learning models on high-dimensional sparse features
• build CTR prediction models for advertising
• streaming training with feature admission and eviction
• serve recommendation recall models with deep tree matching
• scale parameter servers for large sparse embeddings

When to choose

• you train models on very large sparse feature spaces like ads or recommendations
• you need parameter server optimizations and streaming training features
• you want Alibaba's proven industrial deep learning stack

When to avoid

• you need a general-purpose deep learning framework for dense data like images or text
• you want a framework with a large active community and frequent updates
• you need first-class GPU ecosystem support or modern tooling

Facets

framework · maturity maintenance

deep-learning machine-learning llm-training deep-learning machine-learning cpp python sparse-data recommendation-systems parameter-server streaming-training industrial-ml advertising recommendation linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="alibaba/x-deeplearning")