# Long-horizon agentic tasks on HLE Our Settings Source: [Long-horizon agentic tasks on HLE Our Settings]() Collection index: [llms.txt]() Data updated: 2026-03-31T09:42:17.380Z Evaluation of model performance on long-horizon agentic tasks using the HLE Our Settings configuration, measured by Pass@1. 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 | Long-horizon agentic tasks | | Dataset (including any stated split) | HLE Our Settings | | Metric | Pass@1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 - [Pass@1](): Higher is better; unit: PERCENT. ## Recorded results | Method | Pass@1 | Result date | Report paper | | --- | --- | --- | --- | | AgentSwing | 44.4 | Not recorded | 69cb555c0636e5bf4afca207 | | DeepSeek-v3.2 | 43.5 | Not recorded | 69cb555c0636e5bf4afca207 | | DeepSeek-v3.2 | 42 | Not recorded | 69cb555c0636e5bf4afca207 | | DeepSeek-v3.2 | 40.2 | Not recorded | 69cb555c0636e5bf4afca207 | | DeepSeek-v3.2 | 39.6 | Not recorded | 69cb555c0636e5bf4afca207 | | AgentSwing | 35.1 | Not recorded | 69cb555c0636e5bf4afca207 | | GPT-OSS-120B | 34.4 | Not recorded | 69cb555c0636e5bf4afca207 | | GPT-OSS-120B | 34.2 | Not recorded | 69cb555c0636e5bf4afca207 | | GPT-OSS-120B | 34.1 | Not recorded | 69cb555c0636e5bf4afca207 | | GPT-OSS-120B | 33.2 | Not recorded | 69cb555c0636e5bf4afca207 | | AgentSwing | 33.1 | Not recorded | 69cb555c0636e5bf4afca207 | | Tongyi-DR-30B-A3B | 32.7 | Not recorded | 69cb555c0636e5bf4afca207 | | Tongyi-DR-30B-A3B | 32.2 | Not recorded | 69cb555c0636e5bf4afca207 | | Tongyi-DR-30B-A3B | 32 | Not recorded | 69cb555c0636e5bf4afca207 | | Tongyi-DR-30B-A3B | 31.7 | Not recorded | 69cb555c0636e5bf4afca207 | ### AgentSwing Pass@1: 44.4. Result ID: 69cb96f447f34ce924b4433f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"DeepSeek-v3.2"}. This record does not assert independent reproduction. ### DeepSeek-v3.2 Pass@1: 43.5. Result ID: 69cb96f447f34ce924b4433e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Summary"}. This record does not assert independent reproduction. ### DeepSeek-v3.2 Pass@1: 42. Result ID: 69cb96f447f34ce924b4433c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Discard-All"}. This record does not assert independent reproduction. ### DeepSeek-v3.2 Pass@1: 40.2. Result ID: 69cb96f447f34ce924b4433b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Baseline (w/o CM)"}. This record does not assert independent reproduction. ### DeepSeek-v3.2 Pass@1: 39.6. Result ID: 69cb96f447f34ce924b4433d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Keep-Last-N"}. This record does not assert independent reproduction. ### AgentSwing Pass@1: 35.1. Result ID: 69cb96f447f34ce924b4433a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"GPT-OSS-120B"}. This record does not assert independent reproduction. ### GPT-OSS-120B Pass@1: 34.4. Result ID: 69cb96f447f34ce924b44339. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Summary"}. This record does not assert independent reproduction. ### GPT-OSS-120B Pass@1: 34.2. Result ID: 69cb96f447f34ce924b44337. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Discard-All"}. This record does not assert independent reproduction. ### GPT-OSS-120B Pass@1: 34.1. Result ID: 69cb96f447f34ce924b44338. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Keep-Last-N"}. This record does not assert independent reproduction. ### GPT-OSS-120B Pass@1: 33.2. Result ID: 69cb96f447f34ce924b44336. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Context Management":"Baseline (w/o CM)"}. This record does not assert independent reproduction. ### AgentSwing Pass@1: 33.1. Result ID: 69cb96f447f34ce924b44344. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Tongyi-DR-30B-A3B"}. This record does not assert independent reproduction. ### Tongyi-DR-30B-A3B Pass@1: 32.7. Result ID: 69cb96f447f34ce924b44341. [Reporting paper]() Recorded evaluation configuration: {"Context Management":"Discard-All"}. This record does not assert independent reproduction. ### Tongyi-DR-30B-A3B Pass@1: 32.2. Result ID: 69cb96f447f34ce924b44342. [Reporting paper]() Recorded evaluation configuration: {"Context Management":"Keep-Last-N"}. This record does not assert independent reproduction. ### Tongyi-DR-30B-A3B Pass@1: 32. Result ID: 69cb96f447f34ce924b44343. [Reporting paper]() Recorded evaluation configuration: {"Context Management":"Summary"}. This record does not assert independent reproduction. ### Tongyi-DR-30B-A3B Pass@1: 31.7. Result ID: 69cb96f447f34ce924b44340. [Reporting paper]() Recorded evaluation configuration: {"Context Management":"Baseline (w/o CM)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.