Back to Archive
AutoBench Run 6 - August 2026
Twenty models scored on 252 questions across ten subject domains. Score, cost, and latency are the per-answer committee rank, billed cost, and generation time. Routers in the AutoBench Router Benchmark are scored against this answer matrix.
Latest
Date
August 29, 2026
Version
2026-08-29
Models
20
New Models
18
Run data
Model | AutoBench | AAI Index |
|---|---|---|
| 4.78 (#1) | 61 (#1) | |
| 4.74 (#2) | 58 (#4) | |
| 4.74 (#3) | 52 (#10) | |
| 4.73 (#4) | 55 (#7) | |
| 4.69 (#5) | 45 (#11) | |
| 4.68 (#6) | 56 (#6) | |
| 4.67 (#7) | 53 (#8) | |
| 4.65 (#8) | 45 (#12) | |
| 4.65 (#9) | 60 (#3) | |
| 4.61 (#10) | 42 (#13) | |
| 4.6 (#11) | 61 (#2) | |
| 4.59 (#12) | 52 (#9) | |
| 4.56 (#13) | 57 (#5) | |
| 4.49 (#14) | 38 (#15) | |
| 4.44 (#15) | 38 (#14) | |
| 4.36 (#16) | 24 (#19) | |
| 4.35 (#17) | 35 (#16) | |
| 4.35 (#18) | 30 (#17) | |
| 4.15 (#19) | 22 (#20) | |
| 4.12 (#20) | 24 (#18) |
