BrowseComp-Plus leaderboard
As of 6 Nov 2026, the top score on BrowseComp-Plus is 57.6% by GPT-5 (OpenAI), ahead of o3 at 50.5%, across 21 model configurations with a published result (BrowseComp-Plus, result dated 8 Aug 2025). BrowseComp questions answered against a fixed 100k-document corpus instead of the live web, so the retriever and the model can be judged apart. Scores shown are on the BM25 baseline retriever, the one column nearly every submission runs. Model coverage stops in late 2025. BrowseComp-Plus is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- GPT-5 holds the top BrowseComp-Plus score at 57.6% (BrowseComp-Plus, 8 Aug 2025).
- o3 is second on BrowseComp-Plus at 50.5% (BrowseComp-Plus, 8 Aug 2025).
- BrowseComp-Plus: BrowseComp questions answered against a fixed 100k-document corpus instead of the live web, so the retriever and the model can be judged apart. Scores shown are on the BM25 baseline retriever, the one column nearly every submission runs. Model coverage stops in late 2025 (BrowseComp-Plus).
- 21 model configurations have a published BrowseComp-Plus result on BenchLeader (BenchLeader, 6 Nov 2026).
- BrowseComp-Plus does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
BrowseComp questions answered against a fixed 100k-document corpus instead of the live web, so the retriever and the model can be judged apart. Scores shown are on the BM25 baseline retriever, the one column nearly every submission runs. Model coverage stops in late 2025.
- Published by
- BrowseComp-Plusdata via BrowseComp-Plus
- Category
- Agents & tools
- Index weight
- Reference only
- Models
- 21
- Data as of
- 6 Nov 2026
BrowseComp-Plus (texttron/BrowseComp-Plus), results dataset Tevatron/BrowseComp-Plus-results.
- 1GPT-557.6%
- 2o350.5%
- 3gpt-oss-120b (high)43.6%
- 4Tongyi DeepResearch 30B-A3B36.9%
- 5GLM 4.733.3%
- 6GLM-4.627.7%
- 7gpt-oss-120b (medium)24.6%
- 8gpt-oss-20b (high)21.4%
- 9Gemini 2.5 Pro19.9%
- 10gpt-oss-20b (medium)16.9%
- 11Gemini 2.5 Flash16.3%
- 12Claude Opus 415.5%
- 13GPT-4.115.3%
- 14Claude Sonnet 414.7%
- 15Kimi K214.6%
| # | ||||
|---|---|---|---|---|
| 1 |  GPT-5 | 57.6% | 57.7 | 2025-08-08 |
| 2 | o3 | 50.5% | 61.0 | 2025-08-08 |
| 3 | gpt-oss-120bhighopen ↗ | 43.6% | 48.9 | 2025-12-31 |
| 4 |  Tongyi DeepResearch 30B-A3Bopen | 36.9% | – | 2025-12-27 |
| 5 |  GLM 4.7open ↗ | 33.3% | 51.7 | 2025-12-26 |
| 6 | GLM-4.6open ↗ | 27.7% | 49.8 | 2025-10-13 |
| 7 | gpt-oss-120bmediumopen ↗ | 24.6% | – | 2025-08-08 |
| 8 | gpt-oss-20bhighopen ↗ | 21.4% | 45.5 | 2025-08-08 |
| 9 |  Gemini 2.5 Pro | 19.9% | 54.5 | 2025-08-08 |
| 10 | gpt-oss-20bmediumopen ↗ | 16.9% | – | 2025-08-08 |
| 11 | Gemini 2.5 Flash | 16.3% | 52.7 | 2025-08-08 |
| 12 |  Claude Opus 4 | 15.5% | 52.6 | 2025-08-08 |
| 13 | GPT-4.1 | 15.3% | 48.5 | 2025-08-08 |
| 14 | Claude Sonnet 4 | 14.7% | 50.5 | 2025-08-08 |
| 15 |  Kimi K2open ↗ | 14.6% | 51.0 | 2025-08-16 |
| 16 | Websailor 32Bopen | 12.1% | – | 2025-09-28 |
| 17 | gpt-oss-120blowopen ↗ | 9.8% | 45.7 | 2025-08-08 |
| 18 |  DeepSeek R1 0528open ↗ | 8.6% | 49.3 | 2025-08-16 |
| 19 | Searchr1 32Bopen | 4.1% | – | 2025-08-08 |
| 20 | gpt-oss-20blowopen ↗ | 4.0% | 45.6 | 2025-08-08 |
| 21 | Qwen3 32Bopen ↗ | 3.6% | 51.4 | 2025-08-08 |
More agents & tools benchmarks · All LLM benchmarks
Related benchmarks
Cite as: BenchLeader, “BrowseComp-Plus leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/browsecomp_plus.
BrowseComp-Plus: questions
- What does BrowseComp-Plus measure?
- BrowseComp questions answered against a fixed 100k-document corpus instead of the live web, so the retriever and the model can be judged apart. Scores shown are on the BM25 baseline retriever, the one column nearly every submission runs. Model coverage stops in late 2025. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on BrowseComp-Plus?
- As of 6 Nov 2026, the highest score on BrowseComp-Plus is 57.6% by GPT-5 (result dated 8 Aug 2025), ahead of o3 at 50.5%. Scores come from BrowseComp-Plus and are re-read daily.
- Which AI model leads BrowseComp-Plus?
- GPT-5 leads BrowseComp-Plus with 57.6% as of 6 Nov 2026 (result dated 8 Aug 2025), ahead of o3 at 50.5%.
- How many models have BrowseComp-Plus results?
- 21 model configurations have a BrowseComp-Plus result on BenchLeader, all taken from BrowseComp-Plus via BrowseComp-Plus.
- Who runs BrowseComp-Plus and how often is it updated?
- BrowseComp-Plus is published by BrowseComp-Plus. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does BrowseComp-Plus count toward the BenchLeader Index?
- No. BrowseComp-Plus is shown for reference but left out of the composite index.