# Expert-Level Question Answering on Humanity's Last Exam Source: [Expert-Level Question Answering on Humanity's Last Exam]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:42.358Z Evaluation of model performance on the Humanity's Last Exam dataset for Expert-Level Question Answering, measured by Accuracy. 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 | Expert-Level Question Answering | | Dataset (including any stated split) | Humanity's Last Exam | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Seed-1.8 | 40.9 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | DeepSeek-V3.2 | 40.8 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | OpenAI-GPT-5-high | 35.2 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Tongyi-DeepResearch 30B | 32.9 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Minimax-M2 | 31.8 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | WebSailor-V2-30B-A3B | 30.6 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | GLM-4.6 | 30.4 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | IterResearch-30B-A3B | 28.8 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Gemini Deep Research | 26.9 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Kimi-Researcher | 26.9 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Claude-4.5-Sonnet | 24.5 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | MiroThinker 8B | 21.5 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | AgentCPM-Explore-4B | 19.1 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | | Merged-Model-4B | 17.4 | Not recorded | 698969e4c97a0b3e3b5dc3a5 | ### Seed-1.8 Accuracy: 40.9. Result ID: 6995ffe45698665ec435db51. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### DeepSeek-V3.2 Accuracy: 40.8. Result ID: 6995ffe45698665ec435db4c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### OpenAI-GPT-5-high Accuracy: 35.2. Result ID: 6995ffe45698665ec435db4e. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### Tongyi-DeepResearch 30B Accuracy: 32.9. Result ID: 6995ffe45698665ec435db53. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. ### Minimax-M2 Accuracy: 31.8. Result ID: 6995ffe45698665ec435db4d. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### WebSailor-V2-30B-A3B Accuracy: 30.6. Result ID: 6995ffe45698665ec435db55. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. ### GLM-4.6 Accuracy: 30.4. Result ID: 6995ffe45698665ec435db4f. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### IterResearch-30B-A3B Accuracy: 28.8. Result ID: 6995ffe45698665ec435db54. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. ### Gemini Deep Research Accuracy: 26.9. Result ID: 6995ffe45698665ec435db4b. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### Kimi-Researcher Accuracy: 26.9. Result ID: 6995ffe45698665ec435db50. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### Claude-4.5-Sonnet Accuracy: 24.5. Result ID: 6995ffe45698665ec435db4a. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Closed-Source"}. This record does not assert independent reproduction. ### MiroThinker 8B Accuracy: 21.5. Result ID: 6995ffe45698665ec435db52. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. ### AgentCPM-Explore-4B Accuracy: 19.1. Result ID: 6995ffe45698665ec435db57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. ### Merged-Model-4B Accuracy: 17.4. Result ID: 6995ffe45698665ec435db56. [Reporting paper]() Recorded evaluation configuration: {"Model Access Type":"Open-Source"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.