A benchmark that evaluates the ability of AI agents to locate difficult-to-find information through web research.
Browse the latest scores, model modes, release dates, and parameter sizes for BrowseComp.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
![]() Kimi K3 Thinking Level · MaxToolsInternet | 91.20 | 2026-07-16 | 2800B | Free Commercial | |
![]() GPT-5.5 Pro Thinking Level · Extra HighToolsInternet | 90.10 | 2026-04-23 | Unknown | Closed | |
![]() GPT-5.4 Pro Thinking Level · HighTools | 89.30 | 2026-03-05 | Unknown | Closed | |
4 | ![]() Gemini 3.1 Pro Preview Thinking Level · HighToolsInternet | 85.90 | 2026-02-20 | Unknown | Closed |
5 | ![]() Claude Mythos Preview Extended ThinkingTools | 84.90 | 2026-04-07 | Unknown | Closed |
6 | ![]() Claude Sonnet 5 Thinking EnabledToolsInternet | 84.70 | 2026-06-30 | Unknown | Closed |
7 | ![]() GPT-5.5 Thinking Level · HighToolsInternet | 84.40 | 2026-04-23 | Unknown | Closed |
8 | ![]() Claude Opus 4.8 Thinking Level · HighToolsInternet | 84.30 | 2026-05-28 | Unknown | Closed |
9 | ![]() Hy3 Thinking Level · HighToolsInternet | 84.20 | 2026-07-06 | 295B | Free Commercial |
10 | ![]() Claude Opus 4.6 Thinking EnabledToolsInternet | 84.00 | 2026-02-05 | Unknown | Closed |
11 | ![]() MiniMax M3 Thinking EnabledToolsInternet | 83.50 | 2026-06-01 | 428B | Non-Commercial |
12 | ![]() DeepSeek-V4-Pro Thinking Level · Extra HighTools | 83.40 | 2026-04-24 | 1600B | Free Commercial |
13 | ![]() Kimi K2.6 Thinking EnabledToolsInternet | 83.20 | 2026-04-20 | 1000B | Free Commercial |
14 | ![]() GPT-5.4 Thinking Level · Extra HighTools | 82.70 | 2026-03-05 | Unknown | Closed |
15 | ![]() DeepSeek-V4-Pro Thinking Level · HighTools | 80.40 | 2026-04-24 | 1600B | Free Commercial |
16 | ![]() GLM 5.1 Thinking EnabledToolsInternet | 79.30 | 2026-03-27 | 75.4B | Free Commercial |
17 | ![]() Opus 4.7 Extended ThinkingTools | 79.30 | 2026-04-16 | Unknown | Closed |
18 | ![]() Qwen3.5-397B-A17B Thinking EnabledToolsInternet | 78.60 | 2026-02-16 | 39.7B | Free Commercial |
19 | ![]() GPT-5.2 Pro Thinking EnabledTools | 77.90 | 2025-12-11 | Unknown | Closed |
20 | ![]() GPT-5.2 Pro Thinking Level · Extra HighTools | 77.90 | 2025-12-11 | Unknown | Closed |
21 | Inkling Thinking EnabledToolsInternet | 77.10 | 2026-07-15 | 975B | Free Commercial |
22 | ![]() MiniMax M2.5 Thinking EnabledTools | 76.30 | 2026-02-12 | 229B | Free Commercial |
23 | ![]() GLM-5 Thinking EnabledTools | 75.90 | 2026-02-11 | 744B | Free Commercial |
24 | ![]() Step 3.7 Flash Thinking EnabledTools | 75.82 | 2026-05-29 | 198B | Free Commercial |
25 | ![]() GPT-Live-1 Thinking Level · HighToolsInternet | 75.20 | 2026-07-08 | Unknown | Closed |
26 | ![]() Claude Sonnet 4.6 Thinking EnabledTools | 74.70 | 2026-02-17 | Unknown | Closed |
27 | ![]() DeepSeek-V4-Flash Thinking Level · Extra HighTools | 73.20 | 2026-04-24 | 284B | Free Commercial |
28 | ![]() Step 3.5 Flash Thinking EnabledTools | 69.00 | 2026-02-02 | 196B | Free Commercial |
29 | ![]() Qwen3.5-397B-A17B Thinking EnabledTools | 69.00 | 2026-02-16 | 39.7B | Free Commercial |
30 | ![]() GPT-5.2 Thinking Level · Extra HighToolsInternet | 65.80 | 2025-12-11 | Unknown | Closed |