# Medical and Health Knowledge on HealthBench Source: [Medical and Health Knowledge on HealthBench]() Collection index: [llms.txt]() Data updated: 2026-06-02T10:46:50.513Z Evaluation of model performance on general medical and health knowledge questions using the HealthBench dataset, 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 | Medical and Health Knowledge | | Dataset (including any stated split) | HealthBench | | 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. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | GLM-4.7 & ANDES (Ours) | 37.2 | Not recorded | 6a1e640adbd7150506f8564a | | Official Instruct Model | 29.58 | Not recorded | 6a1e640adbd7150506f8564a | | GPT-5.2 | 21.92 | Not recorded | 6a1e640adbd7150506f8564a | | Opus-4.6 (1M) | 21.12 | Not recorded | 6a1e640adbd7150506f8564a | | Opus-4.6 | 18.81 | Not recorded | 6a1e640adbd7150506f8564a | | GPT-5.4 (High) | 18.64 | Not recorded | 6a1e640adbd7150506f8564a | | Gemini-3.1-Pro | 18.45 | Not recorded | 6a1e640adbd7150506f8564a | | Opus-4.7 (xHigh) | 16.53 | Not recorded | 6a1e640adbd7150506f8564a | | GLM-4.7 (Scaffold-only) | 7.38 | Not recorded | 6a1e640adbd7150506f8564a | | Base Model (SmolLM3-3B) | 0 | Not recorded | 6a1e640adbd7150506f8564a | | Sonnet-4.5 | 0 | Not recorded | 6a1e640adbd7150506f8564a | | Qwen3-Max | 0 | Not recorded | 6a1e640adbd7150506f8564a | | Kimi-K2-Thinking | 0 | Not recorded | 6a1e640adbd7150506f8564a | | MiniMax-M2.1 | 0 | Not recorded | 6a1e640adbd7150506f8564a | | GPT-5.1-Codex-Max | 0 | Not recorded | 6a1e640adbd7150506f8564a | | MiniMax-M2.5 | 0 | Not recorded | 6a1e640adbd7150506f8564a | | GLM-4.7 (OpenCode) | 0 | Not recorded | 6a1e640adbd7150506f8564a | ### GLM-4.7 & ANDES (Ours) Accuracy: 37.2. Result ID: 6a1eb47dfbc373592c19a290. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Proposed Method","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Official Instruct Model Accuracy: 29.58. Result ID: 6a1eb47dfbc373592c19a280. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Official Instruct","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### GPT-5.2 Accuracy: 21.92. Result ID: 6a1eb47dfbc373592c19a289. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Opus-4.6 (1M) Accuracy: 21.12. Result ID: 6a1eb47dfbc373592c19a28d. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Opus-4.6 Accuracy: 18.81. Result ID: 6a1eb47dfbc373592c19a28b. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### GPT-5.4 (High) Accuracy: 18.64. Result ID: 6a1eb47dfbc373592c19a288. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Gemini-3.1-Pro Accuracy: 18.45. Result ID: 6a1eb47dfbc373592c19a28a. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Opus-4.7 (xHigh) Accuracy: 16.53. Result ID: 6a1eb47dfbc373592c19a28c. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### GLM-4.7 (Scaffold-only) Accuracy: 7.38. Result ID: 6a1eb47dfbc373592c19a28f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Proposed Method","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Base Model (SmolLM3-3B) Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a281. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Zero-Shot","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Sonnet-4.5 Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a282. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Qwen3-Max Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a283. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### Kimi-K2-Thinking Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a284. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### MiniMax-M2.1 Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a285. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### GPT-5.1-Codex-Max Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a286. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### MiniMax-M2.5 Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a287. [Reporting paper]() Recorded evaluation configuration: {"Protocol":"Agent-Post-Trained","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. ### GLM-4.7 (OpenCode) Accuracy: 0. Result ID: 6a1eb47dfbc373592c19a28e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Protocol":"Proposed Method","Backbone":"SmolLM3-3B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.