# Query-level Safety Classification on General-domain Safety Benchmarks Source: [Query-level Safety Classification on General-domain Safety Benchmarks]() Collection index: [llms.txt]() Data updated: 2026-05-29T08:53:09.723Z Evaluation of model performance on general-domain query-level safety classification tasks, reporting F1 score. 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 | Query-level safety classification | | Dataset (including any stated split) | General-domain safety benchmarks Query-Level | | Metric | F1 Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1 Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen3Guard-Gen-8B | 86.78 | Not recorded | 6a191d76dbd7150506f826e5 | | PolyGuard-Qwen-7B | 86.11 | Not recorded | 6a191d76dbd7150506f826e5 | | Qwen3Guard-Gen-8B | 85.49 | Not recorded | 6a191d76dbd7150506f826e5 | | FinGuard | 81 | Not recorded | 6a191d76dbd7150506f826e5 | | Qwen3-8B | 79.18 | Not recorded | 6a191d76dbd7150506f826e5 | | LlamaGuard3-8B | 75.63 | Not recorded | 6a191d76dbd7150506f826e5 | | NemoGuard-8B | 71.74 | Not recorded | 6a191d76dbd7150506f826e5 | ### Qwen3Guard-Gen-8B F1 Score: 86.78. Result ID: 6a195282fbc373592c13c455. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"strict"}. This record does not assert independent reproduction. ### PolyGuard-Qwen-7B F1 Score: 86.11. Result ID: 6a195282fbc373592c13c454. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3Guard-Gen-8B F1 Score: 85.49. Result ID: 6a195282fbc373592c13c456. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"loose"}. This record does not assert independent reproduction. ### FinGuard F1 Score: 81. Result ID: 6a195282fbc373592c13c458. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3-8B F1 Score: 79.18. Result ID: 6a195282fbc373592c13c457. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LlamaGuard3-8B F1 Score: 75.63. Result ID: 6a195282fbc373592c13c452. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### NemoGuard-8B F1 Score: 71.74. Result ID: 6a195282fbc373592c13c453. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.