--- title: "llm-attacks vs IB4LLMs" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/llm-attacks-llm-attacks-vs-zichuan-liu-ib4llms" tools: ["llm-attacks-llm-attacks", "zichuan-liu-ib4llms"] --- # llm-attacks vs IB4LLMs *GraphCanon updated Aug 5, 2026* ## Verdict Pick llm-attacks if llm-attacks: Universal and Transferable Attacks on Aligned Language Models with dependency on FastChat; pick IB4LLMs if iB4LLMs (IBProtector) is an LLM jailbreak defense method using the Information Bottleneck principle to prevent adversarial prompts while preserving key information. [llm-attacks](https://llm-attacks.org/) reports 4.8k GitHub stars, 633 forks, and 69 open issues, last pushed Aug 2, 2024. [IB4LLMs](https://zichuan-liu.github.io/projects/IBProtector/index.html) has 25 stars, 2 forks, and 4 open issues, last pushed Nov 7, 2024. Figures are from public GitHub metadata via [llm-attacks's repository](https://github.com/llm-attacks/llm-attacks) and [IB4LLMs's repository](https://github.com/zichuan-liu/IB4LLMs). | | [llm-attacks](/tools/llm-attacks-llm-attacks.md) | [IB4LLMs](/tools/zichuan-liu-ib4llms.md) | | --- | --- | --- | | Tagline | Universal and Transferable Attacks on Aligned Language Models | Protecting Your LLMs with Information Bottleneck | | Stars | 4,756 | 25 | | Forks | 633 | 2 | | Open issues | 69 | 4 | | Language | Python | Python | | Adopt for | llm-attacks: Universal and Transferable Attacks on Aligned Language Models with dependency on FastChat. | IB4LLMs (IBProtector) is an LLM jailbreak defense method using the Information Bottleneck principle to prevent adversarial prompts while preserving key information. | | Persona | - | - | | Runtime | - | - | | License | MIT | - | | Categories | Evaluation & Observability, LLM Frameworks | Evaluation & Observability, LLM Frameworks | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [llm-attacks](/tools/llm-attacks-llm-attacks.md) | [IB4LLMs](/tools/zichuan-liu-ib4llms.md) | | --- | --- | --- | | Days since push | 732d | 635d | | Open issues (now) | 69 | 4 | | Owner type | Organization | User | | Full report | [trust report](/tools/llm-attacks-llm-attacks/trust.md) | [trust report](/tools/zichuan-liu-ib4llms/trust.md) | ## Shared compatibility - **Python**: [llm-attacks](/tools/llm-attacks-llm-attacks.md) - Python runtime; [IB4LLMs](/tools/zichuan-liu-ib4llms.md) - Python runtime ## Decision facts: llm-attacks - **Adopt for:** llm-attacks: Universal and Transferable Attacks on Aligned Language Models with dependency on FastChat. ## Decision facts: IB4LLMs - **Pricing:** unknown - License information unavailable; specific model pricing not provided. - **Requirements:** Dependencies include datasets==2.14.5, torch==2.1.1, transformers==4.40.1 among others. - **Adopt for:** IB4LLMs (IBProtector) is an LLM jailbreak defense method using the Information Bottleneck principle to prevent adversarial prompts while preserving key information. ## Choose when ### Choose llm-attacks if… - Tags unique to llm-attacks: alignment-testing, attacks, fastchat-dependency, language-models. - When you need to test the robustness of aligned language models specifically using attacks designed for these systems, - More GitHub stars (4.8k vs 25) - visibility, not fit. ### Choose IB4LLMs if… - Pricing: License information unavailable; specific model pricing not provided.. - Requirements: Dependencies include datasets==2.14.5, torch==2.1.1, transformers==4.40.1 among others.. - Tags unique to IB4LLMs: evaluation scripts, finetuning, inference, information bottleneck. - When you need a specialized tool for guarding against jailbreaks in your language models without losing important data. ## When NOT to use llm-attacks - Do not use if you are evaluating generic or unaligned language models without a need for alignment-specific attack testing, - Avoid when FastChat is not used in your project as llm-attacks explicitly depends on it. ## When NOT to use IB4LLMs - If you require a more generalized model protection approach that does not rely strictly on the Information Bottleneck principle. - When your environment lacks support for specific packages like `fschat==0.2.20` which is crucial and cannot be updated due to potential conflicts. ## Common questions ### What is the difference between llm-attacks and IB4LLMs? llm-attacks: Universal and Transferable Attacks on Aligned Language Models. IB4LLMs: Protecting Your LLMs with Information Bottleneck. See the comparison table for live GitHub stats and shared categories. ### When should I choose llm-attacks over IB4LLMs? Choose llm-attacks over IB4LLMs when Tags unique to llm-attacks: alignment-testing, attacks, fastchat-dependency, language-models; When you need to test the robustness of aligned language models specifically using attacks designed for these systems,; More GitHub stars (4.8k vs 25) - visibility, not fit. ### When should I choose IB4LLMs over llm-attacks? Choose IB4LLMs over llm-attacks when Pricing: License information unavailable; specific model pricing not provided.; Requirements: Dependencies include datasets==2.14.5, torch==2.1.1, transformers==4.40.1 among others.; Tags unique to IB4LLMs: evaluation scripts, finetuning, inference, information bottleneck; When you need a specialized tool for guarding against jailbreaks in your language models without losing important data. ### When should I avoid llm-attacks? Do not use if you are evaluating generic or unaligned language models without a need for alignment-specific attack testing, Avoid when FastChat is not used in your project as llm-attacks explicitly depends on it. ### When should I avoid IB4LLMs? If you require a more generalized model protection approach that does not rely strictly on the Information Bottleneck principle. When your environment lacks support for specific packages like `fschat==0.2.20` which is crucial and cannot be updated due to potential conflicts. ### Is llm-attacks or IB4LLMs more popular on GitHub? llm-attacks has more GitHub stars (4,756 vs 25). Stars measure visibility, not whether either tool fits your constraints. ### Are llm-attacks and IB4LLMs open source? Yes - both are open-source projects on GitHub. ### Where can I find alternatives to llm-attacks or IB4LLMs? GraphCanon lists graph-backed alternatives at [llm-attacks alternatives](/tools/llm-attacks-llm-attacks/alternatives) and [IB4LLMs alternatives](/tools/zichuan-liu-ib4llms/alternatives) ([llm-attacks markdown twin](/tools/llm-attacks-llm-attacks/alternatives.md), [IB4LLMs markdown twin](/tools/zichuan-liu-ib4llms/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/llm-attacks-llm-attacks-vs-zichuan-liu-ib4llms.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, llm-attacks or IB4LLMs? llm-attacks: Dormant. IB4LLMs: Dormant. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for llm-attacks and IB4LLMs? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [llm-attacks trust report](/tools/llm-attacks-llm-attacks/trust); [IB4LLMs trust report](/tools/zichuan-liu-ib4llms/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=llm-attacks-llm-attacks`](/api/graphcanon/graph?tool=llm-attacks-llm-attacks) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_