DI-engine provides an OpenDILab framework for reinforcement learning projects.
Open source
24 alternativesTF-Agents
TF-Agents provides a TensorFlow library for contextual bandits and reinforcement learning.
About TF-Agents
TF-Agents is an open-source Python library built on TensorFlow for implementing, deploying and testing contextual bandit and reinforcement learning algorithms. It implements core reinforcement learning elements as agents, each defining a policy for interacting with an environment and how that policy is trained from collected experience. The library includes a Multi-Armed Bandits suite with environments and agents, along with tutorials and end-to-end training examples.
The project states that it is under active development and that interfaces may change at any time. Using the Reverb replay buffer, which the README describes as very common, works only on Linux. Installing versions of TensorFlow or Reverb that the pip dependency check flags as incompatible is done at the user's own risk.
Key features
- Core reinforcement learning algorithms implemented as agents with policy and training logic
- Agents for DQN, DDQN, DDPG, TD3, REINFORCE, PPO and SAC
- Multi-Armed Bandits suite with bandit environments and agents
- Tutorials and ready-to-run examples, including a DQN agent for the Cartpole environment
- Stable and nightly builds published as pip packages
- Optional Reverb replay buffer support installed through a pip extra
Best for: Researchers and developers who want to build, test or modify reinforcement learning or contextual bandit algorithms in TensorFlow.
Good for
- Components are described as well tested and modular, and can be modified and extended
- Supports fast code iteration with good test integration and benchmarking
- Open source
Watch out for
- Under active development, and interfaces may change at any time
- Reverb replay buffer usage works only on Linux
- Installing incompatible TensorFlow or Reverb versions is at the user's own risk
TF-Agents at a glance
- Website
- github.com/tensorflow/agents
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
TF-Agents alternatives
Tools and sites you could use instead, ranked by how closely they match.
TensorFlow Model Optimization provides Keras and TensorFlow tools for model quantization and pruning.
Open source
24 alternatives
Learnergy is a Python library for energy-based machine learning models.
Open source
24 alternatives
Karate Club offers algorithms for unsupervised learning on graph data in Python.
Open source
24 alternatives
OpenMetricLearning provides metric learning pipelines, models, and a model zoo.
Open source
24 alternatives
AutoKeras automates development of deep learning models through a machine learning library.
Open source
24 alternativesPopular in ML Training & Optimization
Well-known listings from the same topic.
Edwardedwardlib.orgDataturedatature.ioAerosolveaerosolve.airbnb.techTogether AItogether.aiOptunaoptuna.orgCatBoostcatboost.ai
TF-Agents FAQ
Which reinforcement learning algorithms are available in TF-Agents?
The README lists DQN, DDQN, DDPG, TD3, REINFORCE, PPO and SAC as currently available algorithms.
Does the Reverb replay buffer work on every operating system?
No. The README states that if Reverb is used for the replay buffer, TF-Agents will only work with Linux.
Where can I find API documentation for TF-Agents?
API documentation for the current stable release is published on tensorflow.org.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.