# Factuality Evaluation on HaluEval QA (500 items) Source: [Factuality Evaluation on HaluEval QA (500 items)]() Collection index: [llms.txt]() Data updated: 2026-05-19T12:42:56.286Z Evaluation of model factuality performance using the specific 500-item subset of the HaluEval dataset, focusing on MC1 and MC2 metrics. 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 | Factuality Evaluation | | Dataset (including any stated split) | HaluEval QA (500 items) | | Metric | MC1 Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 30 of 30 | | 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 - [MC1 Score](): Higher is better; unit: SCALAR. - [MC2 Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | MC1 Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen3-30B-A3B | 86.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | Ministral-3-14B-R | 83 | Not recorded | 6a0bef07dbc836b87d82c992 | | Qwen3-14B | 79.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | LLaMA-30B | 78.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | DeepSeek-R1-Q-32B | 76.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | LLaMA-7B | 75.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | LLaMA-13B | 75.6 | Not recorded | 6a0bef07dbc836b87d82c992 | | GPT-OSS-20B | 67 | Not recorded | 6a0bef07dbc836b87d82c992 | | Llama-3.3-70B | 61 | Not recorded | 6a0bef07dbc836b87d82c992 | | Phi-4-Reasoning | 57.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | Mixtral-8×7B | 55.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | GPT-OSS-120B | 49.2 | Not recorded | 6a0bef07dbc836b87d82c992 | | Gemma-3-4B | 48.2 | Not recorded | 6a0bef07dbc836b87d82c992 | | Gemma-3-27B | 47 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 43.6 | Not recorded | 6a0bef07dbc836b87d82c992 | | Gemma-3-1B | 41.2 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 34.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 24.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 15.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 14.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 14 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 12.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 8.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 5.8 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 5.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 5.2 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 5 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 3.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 2.4 | Not recorded | 6a0bef07dbc836b87d82c992 | | TRACE | 1 | Not recorded | 6a0bef07dbc836b87d82c992 | ### Qwen3-30B-A3B MC1 Score: 86.4. Result ID: 6a0c59f8fbc373592c03ddd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"3.51","Branch":"mix"}. This record does not assert independent reproduction. ### Ministral-3-14B-R MC1 Score: 83. Result ID: 6a0c59f8fbc373592c03ddec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"0.57","Branch":"early"}. This record does not assert independent reproduction. ### Qwen3-14B MC1 Score: 79.4. Result ID: 6a0c59f8fbc373592c03ddd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"2.57","Branch":"mix"}. This record does not assert independent reproduction. ### LLaMA-30B MC1 Score: 78.8. Result ID: 6a0c59f8fbc373592c03dde8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"I(M)":"0.13","Branch":"early"}. This record does not assert independent reproduction. ### DeepSeek-R1-Q-32B MC1 Score: 76.4. Result ID: 6a0c59f8fbc373592c03dde0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"1.18","Branch":"mix"}. This record does not assert independent reproduction. ### LLaMA-7B MC1 Score: 75.8. Result ID: 6a0c59f8fbc373592c03dde4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"I(M)":"0.21","Branch":"early"}. This record does not assert independent reproduction. ### LLaMA-13B MC1 Score: 75.6. Result ID: 6a0c59f8fbc373592c03dde6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"I(M)":"0.16","Branch":"early"}. This record does not assert independent reproduction. ### GPT-OSS-20B MC1 Score: 67. Result ID: 6a0c59f8fbc373592c03ddda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"3.48","Branch":"mix"}. This record does not assert independent reproduction. ### Llama-3.3-70B MC1 Score: 61. Result ID: 6a0c59f8fbc373592c03dde2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"I(M)":"0.39","Branch":"early"}. This record does not assert independent reproduction. ### Phi-4-Reasoning MC1 Score: 57.4. Result ID: 6a0c59f8fbc373592c03ddea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"0.27","Branch":"early"}. This record does not assert independent reproduction. ### Mixtral-8×7B MC1 Score: 55.4. Result ID: 6a0c59f8fbc373592c03ddde. [Reporting paper]() Recorded evaluation configuration: {"I(M)":"1.30","Branch":"mix"}. This record does not assert independent reproduction. ### GPT-OSS-120B MC1 Score: 49.2. Result ID: 6a0c59f8fbc373592c03dddc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"5.66","Branch":"mix"}. This record does not assert independent reproduction. ### Gemma-3-4B MC1 Score: 48.2. Result ID: 6a0c59f8fbc373592c03ddd2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"11.91","Branch":"mix"}. This record does not assert independent reproduction. ### Gemma-3-27B MC1 Score: 47. Result ID: 6a0c59f8fbc373592c03ddd4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"5.18","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 43.6. Result ID: 6a0c59f8fbc373592c03ddd1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Gemma-3-1B","I(M)":"6.54","Branch":"mix"}. This record does not assert independent reproduction. ### Gemma-3-1B MC1 Score: 41.2. Result ID: 6a0c59f8fbc373592c03ddd0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"I(M)":"6.54","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 34.8. Result ID: 6a0c59f8fbc373592c03ddd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Gemma-3-4B","I(M)":"11.91","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 24.8. Result ID: 6a0c59f8fbc373592c03ddd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Gemma-3-27B","I(M)":"5.18","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 15.4. Result ID: 6a0c59f8fbc373592c03dddd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"GPT-OSS-120B","I(M)":"5.66","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 14.4. Result ID: 6a0c59f8fbc373592c03dddf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mixtral-8×7B","I(M)":"1.30","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 14. Result ID: 6a0c59f8fbc373592c03ddeb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Phi-4-Reasoning","I(M)":"0.27","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 12.8. Result ID: 6a0c59f8fbc373592c03dddb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"GPT-OSS-20B","I(M)":"3.48","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 8.4. Result ID: 6a0c59f8fbc373592c03dde3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama-3.3-70B","I(M)":"0.39","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 5.8. Result ID: 6a0c59f8fbc373592c03dded. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Ministral-3-14B-R","I(M)":"0.57","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 5.4. Result ID: 6a0c59f8fbc373592c03dde1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"DeepSeek-R1-Q-32B","I(M)":"1.18","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 5.2. Result ID: 6a0c59f8fbc373592c03dde9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LLaMA-30B","I(M)":"0.13","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 5. Result ID: 6a0c59f8fbc373592c03dde7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LLaMA-13B","I(M)":"0.16","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 3.4. Result ID: 6a0c59f8fbc373592c03ddd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-14B","I(M)":"2.57","Branch":"mix"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 2.4. Result ID: 6a0c59f8fbc373592c03dde5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"LLaMA-7B","I(M)":"0.21","Branch":"early"}. This record does not assert independent reproduction. ### TRACE MC1 Score: 1. Result ID: 6a0c59f8fbc373592c03ddd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-30B-A3B","I(M)":"3.51","Branch":"mix"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.