# Tool Retrieval and Calling on API-Bank Call+Retrieve

Source: [Tool Retrieval and Calling on API-Bank Call+Retrieve](<https://www.sota2.com/research/sota/tool-retrieval-and-calling-on-api-bank-call-plus-retrieve>)

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

Data updated: 2026-04-21T09:05:14.460Z

Evaluation of model performance on the combined Tool Retrieval and Tool Calling task using the API-Bank Call+Retrieve dataset, measured by Task Completion Rate.

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 | Tool Retrieval and Calling |
| Dataset (including any stated split) | API-Bank Call+Retrieve |
| Metric | Task Completion Rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Task Completion Rate](<https://www.sota2.com/research/sota/tool-retrieval-and-calling-on-api-bank-call-plus-retrieve.md?metric=Task%20Completion%20Rate>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Task Completion Rate | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2.5-7B | 26.9 | Not recorded | 68fa7c041e209131cea8be37 |
| Qwen2.5-7B | 23.5 | Not recorded | 68fa7c041e209131cea8be37 |
| Llama3.1-8B | 22.4 | Not recorded | 68fa7c041e209131cea8be37 |
| Qwen2.5-3B | 21.9 | Not recorded | 68fa7c041e209131cea8be37 |
| Llama3.1-8B | 21 | Not recorded | 68fa7c041e209131cea8be37 |
| Qwen2.5-3B | 18.5 | Not recorded | 68fa7c041e209131cea8be37 |
| Llama3.2-3B | 10.9 | Not recorded | 68fa7c041e209131cea8be37 |
| Llama3.2-3B | 9.2 | Not recorded | 68fa7c041e209131cea8be37 |

### Qwen2.5-7B

Task Completion Rate: 26.9. Result ID: 69e73dbf47f34ce924cda79a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"PA-Tool"}. This record does not assert independent reproduction.

### Qwen2.5-7B

Task Completion Rate: 23.5. Result ID: 69e73dbf47f34ce924cda799.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"Base"}. This record does not assert independent reproduction.

### Llama3.1-8B

Task Completion Rate: 22.4. Result ID: 69e73dbf47f34ce924cda79e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"PA-Tool"}. This record does not assert independent reproduction.

### Qwen2.5-3B

Task Completion Rate: 21.9. Result ID: 69e73dbf47f34ce924cda798.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"PA-Tool"}. This record does not assert independent reproduction.

### Llama3.1-8B

Task Completion Rate: 21. Result ID: 69e73dbf47f34ce924cda79d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"Base"}. This record does not assert independent reproduction.

### Qwen2.5-3B

Task Completion Rate: 18.5. Result ID: 69e73dbf47f34ce924cda797.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"Base"}. This record does not assert independent reproduction.

### Llama3.2-3B

Task Completion Rate: 10.9. Result ID: 69e73dbf47f34ce924cda79b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"Base"}. This record does not assert independent reproduction.

### Llama3.2-3B

Task Completion Rate: 9.2. Result ID: 69e73dbf47f34ce924cda79c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be37.md>)

Recorded evaluation configuration: {"Schema configuration":"PA-Tool"}. This record does not assert independent reproduction.

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