# Agent Planning and API Calling on ToolBench Out of the domain

Source: [Agent Planning and API Calling on ToolBench Out of the domain](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain>)

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

Data updated: 2026-04-02T09:16:18.489Z

Evaluation of model performance on agent planning and API calling capabilities using the Out of the domain split of the ToolBench dataset.

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 | Agent Planning and API Calling |
| Dataset (including any stated split) | ToolBench Out of the domain |
| Metric | Plan Accuracy |
| 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

- [Plan Accuracy](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=Plan%20Accuracy>): Higher is better; unit: PERCENT.

- [Action EM](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=Action%20EM>): Higher is better; unit: PERCENT.

- [Easy F1](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=Easy%20F1>): Higher is better; unit: PERCENT.

- [Hard F1](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=Hard%20F1>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [No Hallucination Rate](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=No%20Hallucination%20Rate>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/agent-planning-and-api-calling-on-toolbench-out-of-the-domain.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Plan Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4o | 83.9 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| DeepSeek-V3 | 77.2 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| DeepSeek-R1 | 74.6 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| HiMA-R1 | 73.4 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| HiMA-SFT | 72.3 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| HiMA-SFT-no | 59.7 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| Qwen2.5-14B | 43.9 | Not recorded | 68fa79ab1e209131cea7ecf3 |
| Qwen2.5-32B | 28.6 | Not recorded | 68fa79ab1e209131cea7ecf3 |

### GPT-4o

Plan Accuracy: 83.9. Result ID: 69ce33df47f34ce924b79b47.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DeepSeek-V3

Plan Accuracy: 77.2. Result ID: 69ce33df47f34ce924b79b49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77001e209131cea7066a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DeepSeek-R1

Plan Accuracy: 74.6. Result ID: 69ce33df47f34ce924b79b48.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### HiMA-R1

Plan Accuracy: 73.4. Result ID: 69ce33df47f34ce924b79b4e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

Recorded evaluation configuration: {"Model Parameters":"3B+3B","Thinking Process":"Enabled"}. This record does not assert independent reproduction.

### HiMA-SFT

Plan Accuracy: 72.3. Result ID: 69ce33df47f34ce924b79b4d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

Recorded evaluation configuration: {"Model Parameters":"7B+3B","Thinking Process":"Enabled"}. This record does not assert independent reproduction.

### HiMA-SFT-no

Plan Accuracy: 59.7. Result ID: 69ce33df47f34ce924b79b4c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

Recorded evaluation configuration: {"Model Parameters":"7B+3B","Thinking Process":"Disabled"}. This record does not assert independent reproduction.

### Qwen2.5-14B

Plan Accuracy: 43.9. Result ID: 69ce33df47f34ce924b79b4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Parameters":"14B"}. This record does not assert independent reproduction.

### Qwen2.5-32B

Plan Accuracy: 28.6. Result ID: 69ce33df47f34ce924b79b4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79ab1e209131cea7ecf3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Parameters":"32B"}. This record does not assert independent reproduction.

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