Tencent Hy3 preview places in the top three on 3 benchmarks.
Tencent Apr 2026 Open weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LongBench v2 | 65.4% | #3 of 27 | Model card | |
| CL-bench | 22.8% | #3 of 24 | Official leaderboard | |
| AdvancedIF | 79.5% | #2 of 6 | Model card | |
| LMArena (Elo) | 1,409 | #121 of 333 | Official leaderboard | |
| CL-bench Life | 15.7% | #6 of 16 | Official leaderboard | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 66.3% | #14 of 31 | Model card | |
| Humanity's Last Exam | 30% | #76 of 147 | Model card | |
| WideSearch | 70.2% | #13 of 17 | Model card | |
| BullshitBench V2 | 16% | #95 of 122 | Official leaderboard | |
| Extended NYT Connections | 17.2% | #83 of 97 | Official leaderboard | |
| BullshitBench | 12.7% | #95 of 107 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 87.2% | #54 of 284 | Model card | |
| IMO-AnswerBench | 84.3% | #15 of 37 | Model card | |
| FrontierScience | 70% | #5 of 7 | Model card | |
| Verifying the Verifiers (Project APE CRED benchmark) | 50.5% | #18 of 23 | Official leaderboard | |
| ResearchClawBench | 12.9 | #22 of 23 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 74.4% | #50 of 150 | Model card | |
| Code Arena | 1,357 | #97 of 124 | Official leaderboard | |
| Monthly-SWEBench | 38% | #10 of 12 | Official leaderboard | |
| Hy-SWE Max | 30% | #3 of 3 | Model card | |
| Hy-Backend | 54.7% | #1 of 1 | Model card | |
| Hy-Vibe Bench | 50.5% | #1 of 1 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| PACT | 1,520 | #13 of 26 | Official leaderboard | |
| SnakeBench | 24 | #92 of 183 | Official leaderboard | |
| Terminal-Bench 2.0 | 54.4% | #32 of 61 | Model card | |
| BrowseComp | 67.1% | #49 of 77 | Model card | |
| Claw-Eval Pass^3 | 55% | #8 of 11 | Model card | |
| WildClawBench | 45.3% | #5 of 5 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| FORTE | 49.7% | #20 of 21 | Official leaderboard |
The benchmarks the lab chose to headline, with the scores it gave.