# Jailbreaking on AdvBench 520 Source: [Jailbreaking on AdvBench 520]() Collection index: [llms.txt]() Data updated: 2026-07-03T07:26:49.877Z Evaluation of model susceptibility to jailbreaking prompts using the specific AdvBench 520 subset/version, reporting Attack Success Rate (ASR%). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Jailbreaking | | Dataset (including any stated split) | AdvBench 520 | | Metric | ASR (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [ASR (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | ASR (%) | Result date | Report paper | | --- | --- | --- | --- | | STEER | 96.7 | Not recorded | 6a47420e143ebdcacacc5ce8 | | STEER | 96.5 | Not recorded | 6a47420e143ebdcacacc5ce8 | | STEER | 96.2 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 95.8 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 88.1 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 85.6 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 84.2 | Not recorded | 6a47420e143ebdcacacc5ce8 | | STEER | 82.1 | Not recorded | 6a47420e143ebdcacacc5ce8 | | STEER | 77.3 | Not recorded | 6a47420e143ebdcacacc5ce8 | | STEER | 76.5 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 74.4 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 69.6 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 58.1 | Not recorded | 6a47420e143ebdcacacc5ce8 | | CSRT | 55.8 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 47.5 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 33.8 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 20.2 | Not recorded | 6a47420e143ebdcacacc5ce8 | | GCG | 18.3 | Not recorded | 6a47420e143ebdcacacc5ce8 | ### STEER ASR (%): 96.7. Result ID: 6a47643860b88bb33992cca1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Mistral-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### STEER ASR (%): 96.5. Result ID: 6a47643860b88bb33992ccad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"GLM-4-9B","Iterations":"@8"}. This record does not assert independent reproduction. ### STEER ASR (%): 96.2. Result ID: 6a47643860b88bb33992cca4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Gemma-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 95.8. Result ID: 6a47643860b88bb33992cca3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Gemma-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 88.1. Result ID: 6a47643860b88bb33992cca2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Gemma-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 85.6. Result ID: 6a47643860b88bb33992cca0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Mistral-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 84.2. Result ID: 6a47643860b88bb33992cc9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Mistral-7B","Iterations":"@8"}. This record does not assert independent reproduction. ### STEER ASR (%): 82.1. Result ID: 6a47643860b88bb33992cca7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama-3-8B","Iterations":"@8"}. This record does not assert independent reproduction. ### STEER ASR (%): 77.3. Result ID: 6a47643860b88bb33992ccaa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"DeepSeek-R1","Iterations":"@8"}. This record does not assert independent reproduction. ### STEER ASR (%): 76.5. Result ID: 6a47643860b88bb33992ccb0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 74.4. Result ID: 6a47643860b88bb33992ccab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"GLM-4-9B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 69.6. Result ID: 6a47643860b88bb33992cca5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama-3-8B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 58.1. Result ID: 6a47643860b88bb33992ccae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Iterations":"@8"}. This record does not assert independent reproduction. ### CSRT ASR (%): 55.8. Result ID: 6a47643860b88bb33992cca8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"DeepSeek-R1","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 47.5. Result ID: 6a47643860b88bb33992cca9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"DeepSeek-R1","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 33.8. Result ID: 6a47643860b88bb33992ccaf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Qwen3-8B","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 20.2. Result ID: 6a47643860b88bb33992ccac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"GLM-4-9B","Iterations":"@8"}. This record does not assert independent reproduction. ### GCG ASR (%): 18.3. Result ID: 6a47643860b88bb33992cca6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Target Model":"Llama-3-8B","Iterations":"@8"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.