# Clinical Task Execution on MedAgentBench v2 (val)

Source: [Clinical Task Execution on MedAgentBench v2 (val)](<https://www.sota2.com/research/sota/clinical-task-execution-on-medagentbench-v2-val>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-29T09:53:30.885Z

Evaluation of model performance on Clinical Task Execution using the validation 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 (val) |
| 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](<https://www.sota2.com/research/sota/clinical-task-execution-on-medagentbench-v2-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GRASP | 77 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 73.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 72.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 72.1 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 71.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 70.4 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 68.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 68.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 68.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 68.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 68.3 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 67.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 66.7 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 65.9 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 65.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 65.4 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 65.3 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 63.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 62.9 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 62.9 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 62.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 62.1 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 60 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 58.3 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 57.6 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 55.4 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 54.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 47.6 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 36.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 31.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 25.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 23.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 19.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 17.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 17.8 | Not recorded | 6a191dcfdbd7150506f82a5c |

### GRASP

Accuracy: 77. Result ID: 6a1961e3fbc373592c142f2f.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### GRASP

Accuracy: 73.8. Result ID: 6a1961e3fbc373592c142f44.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### ExpeL

Accuracy: 72.5. Result ID: 6a1961e3fbc373592c142f41.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764a1e209131cea6c2eb.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### SkillX

Accuracy: 72.1. Result ID: 6a1961e3fbc373592c142f43.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48fc27d0172140ec745ee.md>)

[Implementation](<https://github.com/zjunlp/SkillX>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### GRASP

Accuracy: 71.2. Result ID: 6a1961e3fbc373592c142f4b.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### ExpeL

Accuracy: 70.4. Result ID: 6a1961e3fbc373592c142f48.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764a1e209131cea6c2eb.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### GRASP

Accuracy: 68.8. Result ID: 6a1961e3fbc373592c142f3d.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### Seq. Memory

Accuracy: 68.8. Result ID: 6a1961e3fbc373592c142f3f.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### Evo-MedAgent

Accuracy: 68.8. Result ID: 6a1961e3fbc373592c142f49.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e1be6e7d0172140ec7d61f.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### SkillX

Accuracy: 68.8. Result ID: 6a1961e3fbc373592c142f4a.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48fc27d0172140ec745ee.md>)

[Implementation](<https://github.com/zjunlp/SkillX>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### Batch Memory

Accuracy: 68.3. Result ID: 6a1961e3fbc373592c142f40.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### SkillX

Accuracy: 67.5. Result ID: 6a1961e3fbc373592c142f2e.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48fc27d0172140ec745ee.md>)

[Implementation](<https://github.com/zjunlp/SkillX>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### Evo-MedAgent

Accuracy: 66.7. Result ID: 6a1961e3fbc373592c142f42.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e1be6e7d0172140ec7d61f.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### No skills

Accuracy: 65.9. Result ID: 6a1961e3fbc373592c142f3e.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### ExpeL

Accuracy: 65.5. Result ID: 6a1961e3fbc373592c142f2c.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764a1e209131cea6c2eb.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### SkillX

Accuracy: 65.4. Result ID: 6a1961e3fbc373592c142f3c.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48fc27d0172140ec745ee.md>)

[Implementation](<https://github.com/zjunlp/SkillX>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### Evo-MedAgent

Accuracy: 65.3. Result ID: 6a1961e3fbc373592c142f2d.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e1be6e7d0172140ec7d61f.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### No skills

Accuracy: 63.5. Result ID: 6a1961e3fbc373592c142f45.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### Evo-MedAgent

Accuracy: 62.9. Result ID: 6a1961e3fbc373592c142f3b.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e1be6e7d0172140ec7d61f.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### Batch Memory

Accuracy: 62.9. Result ID: 6a1961e3fbc373592c142f47.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### Batch Memory

Accuracy: 62.2. Result ID: 6a1961e3fbc373592c142f2b.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### Seq. Memory

Accuracy: 62.1. Result ID: 6a1961e3fbc373592c142f46.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"GPT-5.4 (low)"}. This record does not assert independent reproduction.

### Seq. Memory

Accuracy: 60. Result ID: 6a1961e3fbc373592c142f38.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### ExpeL

Accuracy: 58.3. Result ID: 6a1961e3fbc373592c142f3a.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764a1e209131cea6c2eb.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### No skills

Accuracy: 57.6. Result ID: 6a1961e3fbc373592c142f29.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### Batch Memory

Accuracy: 55.4. Result ID: 6a1961e3fbc373592c142f39.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### Seq. Memory

Accuracy: 54.2. Result ID: 6a1961e3fbc373592c142f2a.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"gpt-oss-120b"}. This record does not assert independent reproduction.

### No skills

Accuracy: 47.6. Result ID: 6a1961e3fbc373592c142f37.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"Gemini 3.1 Flash Lite"}. This record does not assert independent reproduction.

### GRASP

Accuracy: 36.8. Result ID: 6a1961e3fbc373592c142f36.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### ExpeL

Accuracy: 31.5. Result ID: 6a1961e3fbc373592c142f33.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa764a1e209131cea6c2eb.md>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### Evo-MedAgent

Accuracy: 25.5. Result ID: 6a1961e3fbc373592c142f34.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e1be6e7d0172140ec7d61f.md>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### No skills

Accuracy: 23.8. Result ID: 6a1961e3fbc373592c142f30.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### SkillX

Accuracy: 19.5. Result ID: 6a1961e3fbc373592c142f35.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d48fc27d0172140ec745ee.md>)

[Implementation](<https://github.com/zjunlp/SkillX>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### Seq. Memory

Accuracy: 17.8. Result ID: 6a1961e3fbc373592c142f31.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

Recorded evaluation configuration: {"Model":"DeepSeek V4 Flash"}. This record does not assert independent reproduction.

### Batch Memory

Accuracy: 17.8. Result ID: 6a1961e3fbc373592c142f32.

[Reporting paper](<https://www.sota2.com/research/paper/6a191dcfdbd7150506f82a5c.md>)

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.
