# Clinical Task Execution on MedAgentBench (val)

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

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

Data updated: 2026-05-29T09:53:30.787Z

Evaluation of model performance on Clinical Task Execution using the validation split of the MedAgentBench 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 | Clinical Task Execution |
| Dataset (including any stated split) | MedAgentBench (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-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GRASP | 86.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 86 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 84.6 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 68.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| GRASP | 66.7 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 62.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 55 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 54.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 52.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 51.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 51.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 50.7 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 49.6 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 49 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 48.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 47.9 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 47.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 46.7 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 46.2 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 45.4 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 45 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 42.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| ExpeL | 42.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 42.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Evo-MedAgent | 42.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 41.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 40.6 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 40.1 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 38.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 36.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 36.5 | Not recorded | 6a191dcfdbd7150506f82a5c |
| No skills | 36.4 | Not recorded | 6a191dcfdbd7150506f82a5c |
| SkillX | 33.8 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Seq. Memory | 33 | Not recorded | 6a191dcfdbd7150506f82a5c |
| Batch Memory | 29.2 | Not recorded | 6a191dcfdbd7150506f82a5c |

### GRASP

Accuracy: 86.2. Result ID: 6a1961e3fbc373592c142edb.

[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.

### GRASP

Accuracy: 86. Result ID: 6a1961e3fbc373592c142ec6.

[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: 84.6. Result ID: 6a1961e3fbc373592c142ee2.

[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.

### GRASP

Accuracy: 68.8. Result ID: 6a1961e3fbc373592c142ecd.

[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.

### GRASP

Accuracy: 66.7. Result ID: 6a1961e3fbc373592c142ed4.

[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.

### Evo-MedAgent

Accuracy: 62.5. Result ID: 6a1961e3fbc373592c142ec4.

[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.

### SkillX

Accuracy: 55. Result ID: 6a1961e3fbc373592c142ecc.

[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.

### SkillX

Accuracy: 54.2. Result ID: 6a1961e3fbc373592c142ed3.

[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: 52.5. Result ID: 6a1961e3fbc373592c142ecb.

[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.

### SkillX

Accuracy: 51.5. Result ID: 6a1961e3fbc373592c142ec5.

[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.

### ExpeL

Accuracy: 51.2. Result ID: 6a1961e3fbc373592c142ed1.

[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.

### ExpeL

Accuracy: 50.7. Result ID: 6a1961e3fbc373592c142eca.

[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: 49.6. Result ID: 6a1961e3fbc373592c142ed2.

[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.

### No skills

Accuracy: 49. Result ID: 6a1961e3fbc373592c142ece.

[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: 48.8. Result ID: 6a1961e3fbc373592c142ecf.

[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.

### Batch Memory

Accuracy: 47.9. Result ID: 6a1961e3fbc373592c142ede.

[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.

### ExpeL

Accuracy: 47.5. Result ID: 6a1961e3fbc373592c142ec3.

[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.

### Seq. Memory

Accuracy: 46.7. Result ID: 6a1961e3fbc373592c142edd.

[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: 46.2. Result ID: 6a1961e3fbc373592c142ed7.

[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: 45.4. Result ID: 6a1961e3fbc373592c142ee1.

[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.

### ExpeL

Accuracy: 45. Result ID: 6a1961e3fbc373592c142edf.

[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.

### Seq. Memory

Accuracy: 42.5. Result ID: 6a1961e3fbc373592c142ed6.

[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: 42.5. Result ID: 6a1961e3fbc373592c142ed8.

[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.

### Evo-MedAgent

Accuracy: 42.5. Result ID: 6a1961e3fbc373592c142ed9.

[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.

### Evo-MedAgent

Accuracy: 42.5. Result ID: 6a1961e3fbc373592c142ee0.

[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.

### No skills

Accuracy: 41.8. Result ID: 6a1961e3fbc373592c142ec7.

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

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

### No skills

Accuracy: 40.6. Result ID: 6a1961e3fbc373592c142edc.

[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.

### No skills

Accuracy: 40.1. Result ID: 6a1961e3fbc373592c142ed5.

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

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

### Seq. Memory

Accuracy: 38.5. Result ID: 6a1961e3fbc373592c142ec1.

[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: 36.8. Result ID: 6a1961e3fbc373592c142ec2.

[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: 36.5. Result ID: 6a1961e3fbc373592c142ec9.

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

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

### No skills

Accuracy: 36.4. Result ID: 6a1961e3fbc373592c142ec0.

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

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

### SkillX

Accuracy: 33.8. Result ID: 6a1961e3fbc373592c142eda.

[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.

### Seq. Memory

Accuracy: 33. Result ID: 6a1961e3fbc373592c142ec8.

[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: 29.2. Result ID: 6a1961e3fbc373592c142ed0.

[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.

Page 1 of 1. This is an index page, not the full collection.
