# Open-ended generation on HelloBench (HB) Source: [Open-ended generation on HelloBench (HB)]() Collection index: [llms.txt]() Data updated: 2026-04-14T12:11:55.449Z Evaluation of models performing Open-ended generation on the HelloBench (HB) dataset, measured by HB-A and HB-B scores. 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 | Open-ended generation | | Dataset (including any stated split) | HelloBench (HB) | | Metric | HB-A Score | | 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 - [HB-A Score](): Higher is better; unit: PERCENT. - [HB-B Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | HB-A Score | Result date | Report paper | | --- | --- | --- | --- | | Qwen3-32B | 84 | Not recorded | 69ddca787d0172140ec79f31 | | Claude 3.7 | 83.9 | Not recorded | 69ddca787d0172140ec79f31 | | GPT-4o | 83.7 | Not recorded | 69ddca787d0172140ec79f31 | | Claude 3.5 | 82.9 | Not recorded | 69ddca787d0172140ec79f31 | | DeepWriter-8B + RL-TCER | 82.2 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen3-8B | 81.4 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen2.5-7B-Instruct + SFT + RL-TCER | 81.1 | Not recorded | 69ddca787d0172140ec79f31 | | DeepWriter-8B + RL-EndoR | 79.6 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen2.5-7B-Instruct + SFT + RL-EndoR | 79 | Not recorded | 69ddca787d0172140ec79f31 | | DeepWriter-8B | 77.8 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen2.5-32B-Instruct | 77 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen2.5-7B-Instruct + SFT | 73.7 | Not recorded | 69ddca787d0172140ec79f31 | | LongWriter-8B + RL-TCER | 72.6 | Not recorded | 69ddca787d0172140ec79f31 | | Qwen2.5-7B-Instruct | 72 | Not recorded | 69ddca787d0172140ec79f31 | | LongWriter-8B + RL-EndoR | 70.1 | Not recorded | 69ddca787d0172140ec79f31 | | LongWriter-8B | 68.8 | Not recorded | 69ddca787d0172140ec79f31 | | Llama3.1-8B-Instruct | 45.5 | Not recorded | 69ddca787d0172140ec79f31 | ### Qwen3-32B HB-A Score: 84. Result ID: 69de2eec47f34ce924c6eb18. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Open-source Models"}. This record does not assert independent reproduction. ### Claude 3.7 HB-A Score: 83.9. Result ID: 69de2eec47f34ce924c6eb14. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Closed-source Models"}. This record does not assert independent reproduction. ### GPT-4o HB-A Score: 83.7. Result ID: 69de2eec47f34ce924c6eb12. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Closed-source Models"}. This record does not assert independent reproduction. ### Claude 3.5 HB-A Score: 82.9. Result ID: 69de2eec47f34ce924c6eb13. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Closed-source Models"}. This record does not assert independent reproduction. ### DeepWriter-8B + RL-TCER HB-A Score: 82.2. Result ID: 69de2eec47f34ce924c6eafd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"DeepWriting Dataset"}. This record does not assert independent reproduction. ### Qwen3-8B HB-A Score: 81.4. Result ID: 69de2eec47f34ce924c6eb17. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Open-source Models"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct + SFT + RL-TCER HB-A Score: 81.1. Result ID: 69de2eec47f34ce924c6eb1c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Finetuned Open-source Models","Training Stage":"SFT + RL-TCER"}. This record does not assert independent reproduction. ### DeepWriter-8B + RL-EndoR HB-A Score: 79.6. Result ID: 69de2eec47f34ce924c6eafc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"DeepWriting Dataset"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct + SFT + RL-EndoR HB-A Score: 79. Result ID: 69de2eec47f34ce924c6eb1b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Finetuned Open-source Models","Training Stage":"SFT + RL-EndoR"}. This record does not assert independent reproduction. ### DeepWriter-8B HB-A Score: 77.8. Result ID: 69de2eec47f34ce924c6eafb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"DeepWriting Dataset"}. This record does not assert independent reproduction. ### Qwen2.5-32B-Instruct HB-A Score: 77. Result ID: 69de2eec47f34ce924c6eb16. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Open-source Models"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct + SFT HB-A Score: 73.7. Result ID: 69de2eec47f34ce924c6eb1a. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Finetuned Open-source Models","Training Stage":"SFT"}. This record does not assert independent reproduction. ### LongWriter-8B + RL-TCER HB-A Score: 72.6. Result ID: 69de2eec47f34ce924c6eb00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"LongWriter Dataset"}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct HB-A Score: 72. Result ID: 69de2eec47f34ce924c6eb19. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Finetuned Open-source Models","Training Stage":"Base"}. This record does not assert independent reproduction. ### LongWriter-8B + RL-EndoR HB-A Score: 70.1. Result ID: 69de2eec47f34ce924c6eaff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"LongWriter Dataset"}. This record does not assert independent reproduction. ### LongWriter-8B HB-A Score: 68.8. Result ID: 69de2eec47f34ce924c6eafe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Finetuning Dataset":"LongWriter Dataset"}. This record does not assert independent reproduction. ### Llama3.1-8B-Instruct HB-A Score: 45.5. Result ID: 69de2eec47f34ce924c6eb15. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"General Open-source Models"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.