B.O.B. Leaderboard
Moonshotreleased 16 Jul 2026
Kimi K3
#24 of 9534 of 46 benchmarksConsistent
62.3
B.O.B.
Category profile
- Real use (hard to game)2×54
- Reasoning & knowledge61
- Math67
- Coding71
- Agents & real work53
- Independent indexes77
Reality check
Lab tests
63
Real use (8 tests)
54
Lab tests and real-use tests agree. It does 9 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)
$0.45 in · $15.00 out
Cost per task on the AA index
$2.00
Output speed (Artificial Analysis)
45 tok/s
§Receipts
Every benchmark result
“Beats” = the share of tracked models this one beat on that test. The raw score is its best reported setting.