# Open-domain long-horizon evaluation on BrowseComp-Plus (BCP) (test)

Source: [Open-domain long-horizon evaluation on BrowseComp-Plus (BCP) (test)](<https://www.sota2.com/research/sota/open-domain-long-horizon-evaluation-on-browsecomp-plus-bcp-test>)

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

Data updated: 2026-06-16T09:46:21.937Z

Evaluation of model performance on the Open-domain long-horizon evaluation task using the test split of the BrowseComp-Plus (BCP) 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 | Open-domain long-horizon evaluation |
| Dataset (including any stated split) | BrowseComp-Plus (BCP) (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](<https://www.sota2.com/research/sota/open-domain-long-horizon-evaluation-on-browsecomp-plus-bcp-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| HiMPO | 10.6 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 10.36 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 9.76 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 9.4 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 9.28 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 9.16 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 8.19 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 8.07 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 7.95 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 7.71 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 7.59 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 7.47 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 7.47 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 7.35 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 7.23 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 7.11 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 6.99 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 6.99 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 6.99 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 6.75 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 6.75 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 6.75 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 6.63 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 6.51 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 6.39 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 6.27 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 6.27 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 5.3 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 5.3 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 5.18 | Not recorded | 6a30d905d33dfb23e783d1ad |
| ReAct | 5.06 | Not recorded | 6a30d905d33dfb23e783d1ad |
| HiMPO | 4.7 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SUPO | 2.53 | Not recorded | 6a30d905d33dfb23e783d1ad |
| MEMPO | 2.29 | Not recorded | 6a30d905d33dfb23e783d1ad |
| SFT only | 2.05 | Not recorded | 6a30d905d33dfb23e783d1ad |

### HiMPO

Accuracy: 10.6. Result ID: 6a311b59fbc373592c2de3aa.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 10.36. Result ID: 6a311b59fbc373592c2de3ab.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 9.76. Result ID: 6a311b59fbc373592c2de3ac.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 9.4. Result ID: 6a311b59fbc373592c2de3a8.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 9.28. Result ID: 6a311b59fbc373592c2de3ad.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 9.16. Result ID: 6a311b59fbc373592c2de3ae.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"full ctx."}. This record does not assert independent reproduction.

### SUPO

Accuracy: 8.19. Result ID: 6a311b59fbc373592c2de3a3.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 8.07. Result ID: 6a311b59fbc373592c2de3a4.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 7.95. Result ID: 6a311b59fbc373592c2de39c.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 7.71. Result ID: 6a311b59fbc373592c2de39d.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### SFT only

Accuracy: 7.59. Result ID: 6a311b59fbc373592c2de395.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 7.47. Result ID: 6a311b59fbc373592c2de38c.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SFT only

Accuracy: 7.47. Result ID: 6a311b59fbc373592c2de396.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 7.35. Result ID: 6a311b59fbc373592c2de3a5.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 7.23. Result ID: 6a311b59fbc373592c2de3a1.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 7.11. Result ID: 6a311b59fbc373592c2de3a6.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 6.99. Result ID: 6a311b59fbc373592c2de39a.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 6.99. Result ID: 6a311b59fbc373592c2de39e.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 6.99. Result ID: 6a311b59fbc373592c2de3a7.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"full ctx."}. This record does not assert independent reproduction.

### SFT only

Accuracy: 6.75. Result ID: 6a311b59fbc373592c2de393.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SFT only

Accuracy: 6.75. Result ID: 6a311b59fbc373592c2de397.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 6.75. Result ID: 6a311b59fbc373592c2de39f.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 6.63. Result ID: 6a311b59fbc373592c2de3a0.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"full ctx."}. This record does not assert independent reproduction.

### SFT only

Accuracy: 6.51. Result ID: 6a311b59fbc373592c2de398.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 6.39. Result ID: 6a311b59fbc373592c2de391.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 6.27. Result ID: 6a311b59fbc373592c2de38e.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### SFT only

Accuracy: 6.27. Result ID: 6a311b59fbc373592c2de399.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"full ctx."}. This record does not assert independent reproduction.

### ReAct

Accuracy: 5.3. Result ID: 6a311b59fbc373592c2de38f.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"8k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 5.3. Result ID: 6a311b59fbc373592c2de390.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"16k ctx.","Compression strategy":"mem\_aware"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 5.18. Result ID: 6a311b59fbc373592c2de38d.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### ReAct

Accuracy: 5.06. Result ID: 6a311b59fbc373592c2de392.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Context budget":"full ctx."}. This record does not assert independent reproduction.

### HiMPO

Accuracy: 4.7. Result ID: 6a311b59fbc373592c2de3a9.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### SUPO

Accuracy: 2.53. Result ID: 6a311b59fbc373592c2de3a2.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bec1e209131cea8b552.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### MEMPO

Accuracy: 2.29. Result ID: 6a311b59fbc373592c2de39b.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69a66bce2d603e6de3e3cc04.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

### SFT only

Accuracy: 2.05. Result ID: 6a311b59fbc373592c2de394.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d905d33dfb23e783d1ad.md>)

Recorded evaluation configuration: {"Context budget":"4k ctx.","Compression strategy":"naive\_recency"}. This record does not assert independent reproduction.

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