hy3
53±2
50interval56
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,308
- Last sync
- 33 min33 minutes ago
All sources syncing normally
- Arena · Text· 33 minutes ago
- Arena · WebDev· 33 minutes ago
- Arena · Vision· 33 minutes ago
- Arena · Agent· 33 minutes ago
- GPQA Diamond· 57 minutes ago
- SWE-bench· 57 minutes ago
- Terminal-Bench· 57 minutes ago
- ARC-AGI-2· 57 minutes ago
Filter by org:
Released:
Hy3 preview
49±1
48interval51
hunyuan-hy3-preview
44±2
42interval47
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | Interval | Swing | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | i | Add to compare | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 318 | hy3 | tencent | Jul 2026 | $0.08 / $0.33 | 53.4 56.0Ranked 53.4 = (4 × 56.0 + 3 × 50) ÷ 7. 56.0 is the mean percentile over the 4 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 50–56 | 64 | 4/22 | #65–318 | 84 | 63 | — | 58 | — | — | 20 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | ||
| 432 | Hy3 preview | tencent | Apr 2026 | $0.18 / $0.60 | 49.4 48.5Ranked 49.4 = (2 × 48.5 + 3 × 50) ÷ 5. 48.5 is the mean percentile over the 2 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 48–51 | 47 | 2/22low | #111–432 | — | 25 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72 | — | ||
| 632 | hunyuan-hy3-preview | tencent | Oct 2026 | — | 43.7 37.4Ranked 43.7 = (3 × 37.4 + 3 × 50) ÷ 6. 37.4 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50. | 42–47 | 42 | 3/22 | #116–632 | 65 | 24 | — | 23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |
Blend is coverage-shrunk; the raw composite is shown greyed beside it · A low badge means fewer than 3 boards scored this model