

Apache SINGA
β Text source-checked Nov 10, 2026Apache SINGA is an Apache Top Level Project focusing on distributed training of deep learning and machine learning models.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
In short
Apache SINGA is a free, open-source tool for distributed deep learning training. It supports multi-GPU setups and ONNX model interoperability via a Python interface.
Apache SINGA is an Apache Top Level Project that focuses on the distributed training of deep learning and machine learning models. Apache SINGA supports data parallel training across multiple GPUs on a single node or across different nodes.
Apache SINGA records the computation graph and applies backward propagation automatically after forward propagation. Apache SINGA supports various popular optimizers including stochastic gradient descent with momentum, Adam, RMSProp, and AdaGrad. Apache SINGA supports loading ONNX format models and saving models defined using SINGA APIs into ONNX format to enable use across different libraries and tools.
Apache SINGA provides a well-architected software stack and an easy-to-use Python interface to improve usability. Apache SINGA works with Python 3.9, 3.10, and 3.11. Apache SINGA offers easy installation using Conda, Pip, Docker, and from source.
According to Apache SINGA: Apache SINGA is an Apache Top Level Project that focuses on the distributed training of deep learning and machine learning models. Its pages list this strength: Supports data parallel training across multiple GPUs on single or multiple nodes.
β The AI Tool Bible editorial team
Pros
- β Supports data parallel training across multiple GPUs on single or multiple nodes.
- β Automatically records computation graphs and applies backward propagation.
- β Supports loading and saving models in ONNX format for interoperability.
- β Provides an easy-to-use Python interface.
Facts
- Platforms
- api (not source-checked)
- Pricing
- free - Free, Apache 2.0 licensed (not source-checked)
- API available
- Yes (not source-checked)
- Open source
- Yes (license: Apache-2.0) (not source-checked)
- GitHub stars
- 3,604 (checked Nov 13, 2026) (not source-checked)
- Last GitHub push
- Oct 30, 2026 (not source-checked)
- First commit
- Apr 2015 (not source-checked)
- Company
- Apache Software Foundation (not source-checked)
- Hacker News
- 5 stories Β· top 3 points Β· last Apr 20, 2020 Β· checked Nov 7, 2026
Where this tool sits
Use cases
Frequently asked
- How much does Apache SINGA cost?
- Apache SINGA is completely free to use. It is released under the Apache 2.0 license, making it an open-source option for distributed deep learning training without licensing fees.
- Can I import models from other frameworks?
- Yes, Apache SINGA supports loading models in the ONNX format. It can also save models defined using SINGA APIs into ONNX format, enabling interoperability across different libraries and tools.
- What Python versions are supported?
- Apache SINGA is compatible with Python 3.9, 3.10, and 3.11. It provides an easy-to-use Python interface designed to improve usability for developers working with these specific versions.
- How do I install Apache SINGA?
- Installation is straightforward using Conda, Pip, Docker, or by building from source. This flexibility allows developers to integrate the tool into their preferred development environments easily.
- Does it support distributed training across multiple nodes?
- Yes, Apache SINGA supports data parallel training across multiple GPUs on a single node or across different nodes. This capability is central to its focus on distributed training of deep learning models.