# Vulnerability Prediction on CWE-Prediction Source: [Vulnerability Prediction on CWE-Prediction]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:45.381Z Evaluation of a model's ability to predict vulnerabilities using the CWE-Prediction dataset, measured by Accuracy and Standard Deviation. 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 | Vulnerability Prediction | | Dataset (including any stated split) | CWE-Prediction | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [Std Dev](): Lower is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Foundation-Sec-8B-Reasoning | 70.4 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-5 | 70.1 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-4.1 | 68.2 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-5-Mini | 66.6 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-OSS-120B | 66.1 | Not recorded | 697c3b39a8b01e76cc8da5bc | | O3-Mini | 63.2 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Foundation-Sec-8B-Instruct | 61.6 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Llama-Primus-Nemotron-70B-Instruct | 61.3 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-5-Nano | 61.2 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Llama-3.3-70B-Instruct | 60.7 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Phi-4 | 55.4 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Llama-Primus-Merged | 55 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Qwen-3-14B | 53.4 | Not recorded | 697c3b39a8b01e76cc8da5bc | | GPT-OSS-20B | 51.9 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Llama-3.1-8B-Instruct | 47.3 | Not recorded | 697c3b39a8b01e76cc8da5bc | | Qwen-3-8B | 46.4 | Not recorded | 697c3b39a8b01e76cc8da5bc | | DeepHat-V1-7B | 36 | Not recorded | 697c3b39a8b01e76cc8da5bc | ### Foundation-Sec-8B-Reasoning Accuracy: 70.4. Result ID: 69961c155698665ec45c34a8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 70.1. Result ID: 69961c155698665ec45c34a7. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-4.1 Accuracy: 68.2. Result ID: 69961c155698665ec45c34a3. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-5-Mini Accuracy: 66.6. Result ID: 69961c155698665ec45c34a6. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-OSS-120B Accuracy: 66.1. Result ID: 69961c155698665ec45c34a2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### O3-Mini Accuracy: 63.2. Result ID: 69961c155698665ec45c34a4. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Foundation-Sec-8B-Instruct Accuracy: 61.6. Result ID: 69961c155698665ec45c349e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Llama-Primus-Nemotron-70B-Instruct Accuracy: 61.3. Result ID: 69961c155698665ec45c34a0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-5-Nano Accuracy: 61.2. Result ID: 69961c155698665ec45c34a5. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Llama-3.3-70B-Instruct Accuracy: 60.7. Result ID: 69961c155698665ec45c349d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Phi-4 Accuracy: 55.4. Result ID: 69961c155698665ec45c349b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Llama-Primus-Merged Accuracy: 55. Result ID: 69961c155698665ec45c349f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Qwen-3-14B Accuracy: 53.4. Result ID: 69961c155698665ec45c349a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### GPT-OSS-20B Accuracy: 51.9. Result ID: 69961c155698665ec45c34a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Llama-3.1-8B-Instruct Accuracy: 47.3. Result ID: 69961c155698665ec45c349c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### Qwen-3-8B Accuracy: 46.4. Result ID: 69961c155698665ec45c3499. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. ### DeepHat-V1-7B Accuracy: 36. Result ID: 69961c155698665ec45c3498. [Reporting paper]() Recorded evaluation configuration: {"Number of trials":5}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.