HarmBench vs HELM
HarmBench
github.com
HELM
crfm.stanford.eduSide by side
| Figure | HarmBench | HELM |
|---|---|---|
| Price from | — | Free |
| Free plan | — | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | — | Yes |
| Linux | Yes | — |
| API | Yes | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 0 | 1 |
| Spec score · LLM Evaluation Tools | 7.2 (#21) | 8.3 (#7) |
| Custom models | Users can add Hugging Face Transformers models through the model configuration file, though AutoDAN, PAIR, and TAP require manual experiment configuration for new models. | Users can configure private models, including a checkpoint on local disk or a model server on a private network. |
| Installation | The README instructs users to clone the repository, install requirements with pip, and download the en_core_web_sm spaCy model. | The project instructs users to install HELM from PyPI with `pip install crfm-helm`. |
| License | The public GitHub repository lists an MIT license. | The repository identifies HELM as open source and links to an Apache-2.0 license. |
| Purpose | HarmBench is an open-source framework for evaluating automated red teaming methods and LLM attacks and defenses. | HELM is an open-source Python framework for holistic, reproducible, and transparent evaluation of foundation models, including language and multimodal models. |
| Support | The repository documentation invites users to submit a pull request for support for another cluster manager. | Users are directed to open GitHub issues for questions, which maintainers address when bandwidth is available. |
Plans
In their own words
HELM
HELM is an open-source Python framework for evaluating foundation models, including large language and multimodal systems. It supplies standardized datasets and benchmarks such as MMLU-Pro, GPQA, IFEval, and WildBench, with a unified interface for models from providers including OpenAI, Anthropic, and Google. Evaluations can examine accuracy, efficiency, bias, and toxicity, and can include custom metrics, LLM-as-a-judge checks, safety evaluations, and human review workflows. Users can configure private checkpoints or model servers, run inference locally, or send requests to an API. A web interface displays individual prompts and responses, while leaderboards compare models and benchmarks across areas such as medicine, finance, multilinguality, and world knowledge. HELM can reproduce published evaluation results and documents commands for running benchmarks, summarizing results, and starting a local server. It is free, Apache-2.0 licensed, and entered maintenance mode on June 1, 2026.