Horovod
horovod.ai
- Free plan
- macOS
- Linux
- API
- Self-hosted
7.3 Club score Visit horovod.ai
Plans
Every plan and priceHorovod
Free
Free plan
What the maker publishes
- Autotuning
- Horovod includes automated performance tuning that can be enabled with a command-line argument. source
- Collectives
- Its distributed optimizer averages gradients using allreduce or allgather before applying them. source
- Deployment
- Horovod can be installed on-premise or run on AWS, Azure and Databricks. source
- Installation
- The getting-started page provides pip installation commands for CPU use and GPU use with NCCL. source
- MPI alternative
- Gloo is included and allows Horovod to run without MPI installed. source
- Open source
- Horovod is described as an open source project hosted by the LF AI & Data Foundation. source
- Performance
- Tensor Fusion batches small allreduce operations and interleaves communication with computation. source
- Project status
- The repository says Horovod was archived in September 2026 due to inactivity and is read-only. source
- Purpose
- Horovod is a distributed deep learning training framework for TensorFlow, Keras, PyTorch and Apache MXNet. source
- Requirements
- The project README says Linux or macOS installation requires CMake and describes compiler requirements for supported frameworks. source
- Runtime integrations
- Documentation describes running Horovod with Docker, Kubernetes, Spark, Ray, Singularity, LSF and Hadoop Yarn. source
- Scale
- The project site says an existing training script can be scaled to hundreds of GPUs with a few lines of Python code. source
- Security
- The documentation includes a security section for Horovod on Spark; the opened pages make no security certification or compliance claim. source
- Spark
- Horovod can run on Apache Spark to combine data processing and model training in one pipeline. source
- Support
- The project directs users to documentation, troubleshooting guidance and a ticket submission link for unresolved issues. source
Checked against the maker's pages on 9 Oct 2026.