EasyJailbreak vs promptbench
LLM jailbreak research and model evaluation compared
EasyJailbreak and promptbench are Python projects for studying language model behavior. EasyJailbreak centers on assembling and evaluating jailbreak attacks, while promptbench covers broader benchmark, prompt, robustness, dynamic, and multimodal evaluation.

EasyJailbreak: Build and Evaluate LLM Jailbreak Attacks
EasyJailbreak is a Python framework for assembling and testing jailbreak methods against language models. It suits researchers and developers who need reusable attack components and a structured way to evaluate model responses.

promptbench: Evaluate LLMs and Test Prompt Robustness
PromptBench is a Python library for evaluating language and multimodal models across datasets, prompting methods, and adversarial attacks. It suits researchers and developers comparing model behavior or studying robustness and dynamic evaluation.
| EasyJailbreak | promptbench | |
|---|---|---|
| Language | Python | Python |
| License | GPL-3.0 | MIT |
| Stars | 914 | 2.8k |
| Forks | 92 | 222 |
| Last analyzed | Oct 3, 2026 | Oct 3, 2026 |
Key differences
- EasyJailbreak organizes jailbreak experiments around prompt selection, mutation, constraints, model inference, and evaluation; promptbench provides shared APIs for models, datasets, prompting methods, and evaluation protocols.
- EasyJailbreak includes attack recipes such as PAIR, GCG, GPTFuzz, AutoDAN, and TAP; promptbench includes prompting techniques and adversarial attacks at character, word, sentence, and semantic levels.
- EasyJailbreak focuses on jailbreak research with language models; promptbench also covers multimodal evaluation and dynamic sample generation through DyVal.
- EasyJailbreak uses the GPL-3.0 license; promptbench uses the MIT license.
- EasyJailbreak is not archived; promptbench is archived, so its maintenance status is worth checking before relying on it for ongoing work.
- EasyJailbreak has 914 stars and 92 forks; promptbench has 2.8k stars and 222 forks.
Choose EasyJailbreak if you…
- need reusable components for assembling or adapting jailbreak attacks.
- want to compare included jailbreak recipes against selected target models.
- plan to produce reports with prompts, model responses, and evaluation scores.
Choose promptbench if you…
- need to compare models across supported datasets and evaluation protocols.
- want to study prompting methods, adversarial robustness, or attack transferability.
- need dynamic evaluation or supported image-and-language benchmarks.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.