# Clinical Task Execution on MedAgentBench v2 (test) Source: [Clinical Task Execution on MedAgentBench v2 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-29T09:53:30.914Z Evaluation of model performance on executing clinical tasks using the test split of MedAgentBench version 2, 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 | Clinical Task Execution | | Dataset (including any stated split) | MedAgentBench v2 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 35 of 35 | | 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 | | --- | --- | --- | --- | | GRASP | 76.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | SkillX | 74 | Not recorded | 6a191dcfdbd7150506f82a5c | | Batch Memory | 73.4 | Not recorded | 6a191dcfdbd7150506f82a5c | | ExpeL | 72.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | No skills | 70.5 | Not recorded | 6a191dcfdbd7150506f82a5c | | GRASP | 70.3 | Not recorded | 6a191dcfdbd7150506f82a5c | | Evo-MedAgent | 69.3 | Not recorded | 6a191dcfdbd7150506f82a5c | | ExpeL | 67.8 | Not recorded | 6a191dcfdbd7150506f82a5c | | Seq. Memory | 67.7 | Not recorded | 6a191dcfdbd7150506f82a5c | | GRASP | 67.2 | Not recorded | 6a191dcfdbd7150506f82a5c | | Evo-MedAgent | 66.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | SkillX | 65.1 | Not recorded | 6a191dcfdbd7150506f82a5c | | SkillX | 64.1 | Not recorded | 6a191dcfdbd7150506f82a5c | | Batch Memory | 63.5 | Not recorded | 6a191dcfdbd7150506f82a5c | | GRASP | 63.3 | Not recorded | 6a191dcfdbd7150506f82a5c | | Evo-MedAgent | 62.5 | Not recorded | 6a191dcfdbd7150506f82a5c | | No skills | 61.1 | Not recorded | 6a191dcfdbd7150506f82a5c | | Seq. Memory | 60.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | ExpeL | 60.4 | Not recorded | 6a191dcfdbd7150506f82a5c | | No skills | 59.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | Batch Memory | 59.4 | Not recorded | 6a191dcfdbd7150506f82a5c | | SkillX | 57.8 | Not recorded | 6a191dcfdbd7150506f82a5c | | ExpeL | 55.7 | Not recorded | 6a191dcfdbd7150506f82a5c | | Evo-MedAgent | 54.2 | Not recorded | 6a191dcfdbd7150506f82a5c | | Seq. Memory | 53.8 | Not recorded | 6a191dcfdbd7150506f82a5c | | Batch Memory | 51.6 | Not recorded | 6a191dcfdbd7150506f82a5c | | Seq. Memory | 50.5 | Not recorded | 6a191dcfdbd7150506f82a5c | | No skills | 44.6 | Not recorded | 6a191dcfdbd7150506f82a5c | | GRASP | 30.9 | Not recorded | 6a191dcfdbd7150506f82a5c | | Evo-MedAgent | 25.3 | Not recorded | 6a191dcfdbd7150506f82a5c | | ExpeL | 22.8 | Not recorded | 6a191dcfdbd7150506f82a5c | | No skills | 20.4 | Not recorded | 6a191dcfdbd7150506f82a5c | | Batch Memory | 16.6 | Not recorded | 6a191dcfdbd7150506f82a5c | | SkillX | 14.7 | Not recorded | 6a191dcfdbd7150506f82a5c | | Seq. Memory | 13.8 | Not recorded | 6a191dcfdbd7150506f82a5c | ### GRASP Accuracy: 76.9. Result ID: 6a1961e3fbc373592c142f52. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### SkillX Accuracy: 74. Result ID: 6a1961e3fbc373592c142f66. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### Batch Memory Accuracy: 73.4. Result ID: 6a1961e3fbc373592c142f63. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### ExpeL Accuracy: 72.9. Result ID: 6a1961e3fbc373592c142f64. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### No skills Accuracy: 70.5. Result ID: 6a1961e3fbc373592c142f61. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### GRASP Accuracy: 70.3. Result ID: 6a1961e3fbc373592c142f67. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### Evo-MedAgent Accuracy: 69.3. Result ID: 6a1961e3fbc373592c142f65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### ExpeL Accuracy: 67.8. Result ID: 6a1961e3fbc373592c142f4f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### Seq. Memory Accuracy: 67.7. Result ID: 6a1961e3fbc373592c142f62. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction. ### GRASP Accuracy: 67.2. Result ID: 6a1961e3fbc373592c142f60. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### Evo-MedAgent Accuracy: 66.9. Result ID: 6a1961e3fbc373592c142f50. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### SkillX Accuracy: 65.1. Result ID: 6a1961e3fbc373592c142f5f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### SkillX Accuracy: 64.1. Result ID: 6a1961e3fbc373592c142f51. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### Batch Memory Accuracy: 63.5. Result ID: 6a1961e3fbc373592c142f6a. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### GRASP Accuracy: 63.3. Result ID: 6a1961e3fbc373592c142f6e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### Evo-MedAgent Accuracy: 62.5. Result ID: 6a1961e3fbc373592c142f6c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### No skills Accuracy: 61.1. Result ID: 6a1961e3fbc373592c142f4c. [Reporting paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### Seq. Memory Accuracy: 60.9. Result ID: 6a1961e3fbc373592c142f69. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### ExpeL Accuracy: 60.4. Result ID: 6a1961e3fbc373592c142f6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### No skills Accuracy: 59.9. Result ID: 6a1961e3fbc373592c142f68. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### Batch Memory Accuracy: 59.4. Result ID: 6a1961e3fbc373592c142f4e. [Reporting paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### SkillX Accuracy: 57.8. Result ID: 6a1961e3fbc373592c142f6d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction. ### ExpeL Accuracy: 55.7. Result ID: 6a1961e3fbc373592c142f5d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### Evo-MedAgent Accuracy: 54.2. Result ID: 6a1961e3fbc373592c142f5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### Seq. Memory Accuracy: 53.8. Result ID: 6a1961e3fbc373592c142f4d. [Reporting paper]() Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction. ### Batch Memory Accuracy: 51.6. Result ID: 6a1961e3fbc373592c142f5c. [Reporting paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### Seq. Memory Accuracy: 50.5. Result ID: 6a1961e3fbc373592c142f5b. [Reporting paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### No skills Accuracy: 44.6. Result ID: 6a1961e3fbc373592c142f5a. [Reporting paper]() Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction. ### GRASP Accuracy: 30.9. Result ID: 6a1961e3fbc373592c142f59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### Evo-MedAgent Accuracy: 25.3. Result ID: 6a1961e3fbc373592c142f57. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### ExpeL Accuracy: 22.8. Result ID: 6a1961e3fbc373592c142f56. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### No skills Accuracy: 20.4. Result ID: 6a1961e3fbc373592c142f53. [Reporting paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### Batch Memory Accuracy: 16.6. Result ID: 6a1961e3fbc373592c142f55. [Reporting paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### SkillX Accuracy: 14.7. Result ID: 6a1961e3fbc373592c142f58. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. ### Seq. Memory Accuracy: 13.8. Result ID: 6a1961e3fbc373592c142f54. [Reporting paper]() Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.