Tencent
availableShows if the model has enough results for an index.Hy3 Preview
Hy3 Preview is a reasoning model from Tencent in the Hy3 family. 16 benchmarks count toward its score, in 5 categories.
IndexOverall score out of 100.55.6 ±6.5
CoverageShare of the index weight with results.75%
SpeedOutput tokens per second.86/s
Input / 1MUS dollars per 1M input tokens.$0.18
Output / 1MUS dollars per 1M output tokens.$0.6
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.N/A
The index is a score out of 100. The ± range shows how much it can change.
CapabilitiesScore per category, out of 100.
Out of 100Results
16 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result as a score out of 100. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| Artificial Analysis GPQA Diamond | Knowledge | 89.7% | 60.5 | — | — | Artificial Analysis |
| Graduate-Level Google-Proof Q&A | Knowledge | 87.2% | 58.7 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 87.2% | 58.7 | — | — | David Rein et al. |
| Software Engineering Benchmark Verified | Coding | 74.4% | 57.2 | — | — | Carlos E. Jimenez et al. |
| Artificial Analysis Long Context Reasoning | Reasoning | 66.7% | 54.4 | — | — | Artificial Analysis |
| Instruction Following Benchmark | Instruction | 63.1% | 39.5 | — | — | Benchmark authors |
| Artificial Analysis Coding Index | Coding | 58.8% | 60.4 | — | — | Artificial Analysis |
| Artificial Analysis SciCode | Coding | 48.6% | 60.3 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 41.2% | 74.0 | — | — | Artificial Analysis |
| Scientific Code Benchmark | Coding | 41.2% | 51.2 | — | — | Benchmark authors |
| Gert Labs Composite Game Benchmark | Agentic | 36.9% | 45.3 | — | — | Gert Labs |
| GDPval-AA normalized | Agentic | 35.8% | 63.1 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 33.5% | 60.9 | — | — | Artificial Analysis |
| Artificial Analysis Omniscience Accuracy | Knowledge | 31.5% | 52.8 | — | — | Artificial Analysis |
| Artificial Analysis Agentic Index | Agentic | 25.6% | 58.1 | — | — | Artificial Analysis |
| Humanity's Last Exam | Knowledge | 25.5% | 50.4 | — | — | Center for AI Safety et al. |
| Critical Physics Tasks | Reasoning | 4.9% | 48.7 | — | — | Artificial Analysis |
16 benchmarks count, from 17 of 17 results. A grey row does not count. Too few models took that benchmark.