B.O.B. Leaderboard
OpenAIreleased 3 Sep 2026
GPT-6 Astra
#2 of 9534 of 46 benchmarksConsistent
94.2
B.O.B.
Category profile
- Real use (hard to game)2×91
- Reasoning & knowledge99
- Math96
- Coding94
- Agents & real work93
- Independent indexes96
Reality check
Lab tests
95
Real use (7 tests)
91
Lab tests and real-use tests agree. It does 4 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)
$10.00 in · $50.00 out
Cost per task on the AA index
$3.26
Output speed (Artificial Analysis)
57 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.