FuzzBench is a large-scale, open research platform developed by Google to evaluate and benchmark fuzzers — automated software testing tools that detect vulnerabilities through randomized input generation. It provides a standardized, reproducible environment for comparing the performance and effectiveness of different fuzzing algorithms on real-world software targets. FuzzBench integrates with the OSS-Fuzz infrastructure, allowing it to run experiments on authentic open source projects and collect meaningful data on crash discovery rates, code coverage, and bug-finding efficiency. The service includes an easy-to-use API for integrating custom fuzzers and an automated reporting system that generates detailed statistical analyses, comparative graphs, and significance testing. By running experiments at Google scale, FuzzBench ensures consistent, unbiased, and data-driven evaluations that support academic and industrial fuzzing research.
Features
- Cloud-based evaluation framework for large-scale fuzzer benchmarking
- Uses real-world OSS-Fuzz projects as benchmarks for realistic testing
- Provides statistical reports with visualizations and significance tests
- Simple API for integrating and testing custom fuzzers
- Automatically generates reproducible, large-scale experiments and reports
- Supports open collaboration and contribution from the fuzzing research community