B.O.B. Leaderboard
Tencentreleased 28 Aug 2026
Tencent Hy4 preview
Not ranked yet4 of 46 benchmarksBenchmaxxed?
—
B.O.B.
Not ranked yet: B.O.B. needs at least 5 benchmarks across 3 categories. This model has 4 so far.
Category profile
- Real use (hard to game)2×59
- Reasoning & knowledge—
- Math79
- Coding—
- Agents & real work—
- Independent indexes—
Reality check
Lab tests
79
Real use (3 tests)
59
Scores much higher on lab tests than on real-use tests. It does 20 points better on the usual lab tests than on tests that are hard to train for.
What counts as real usePrice per 1M tokens (OpenRouter)
$0.83 in · $2.50 out
Cost per task on the AA index
—
Output speed (Artificial Analysis)
—
§Receipts
Every benchmark result
“Beats” = the share of tracked models this one beat on that test. The raw score is its best reported setting.
| Benchmark | Raw | Beats | Source |
|---|---|---|---|
| Real use (hard to game) · Real use 59 | |||
| LMArena WebDevas hy4-preview | 1632 | 82%of 82 | LMArena (blind human votes) |
| LMArena Agent Arenaas Hy4 preview | 0.029 | 61%of 50 | LMArena (real agent sessions) |
| Vals Indexas tencent/hy4-preview | 49.9% | 34%of 42 | Vals AI (private test sets) |
| Math · Math 79 | |||
| ProofBenchas hy4-preview_unknown | 75.0% | 79%of 68 | Epoch AI Benchmarking Hub |