Swiss AI Initiative
Apertus 8B Instruct
Apertus 8B Instruct is a Swiss AI Initiative language-model profile in the Easy Benchmarks snapshot. Use this page to compare its measured Artificial Analysis scores, output speed, time to first token, pricing, and relative ranking against other models in the local catalog.
Swiss AI Initiative updatesOperational Metrics
Output Speedn/a
First Tokenn/a
Blended Price$0.125/M
Model Metadata
Queryable facts extracted from the upstream model payload.
ReleaseSep 2, 2025
Context Windown/a
Modalitiesn/a
Strength: Output $
Rank #21 across 325 models.
$0.200/M
Strength: Blended $
Rank #26 across 325 models.
$0.125/M
Strength: Input $
Rank #27 across 325 models.
$0.100/M
Watch Area: Overall
Rank #495 across 500 models.
5.9
Watch Area: GPQA
Rank #462 across 478 models.
25.6%
Watch Area: Coding
Rank #392 across 410 models.
1.4
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 5.9 | #495 |
| Artificial Analysis Coding Index | coding | 1.4 | #392 |
| GPQA | reasoning | 25.6% | #462 |
| Humanity's Last Exam | reasoning | 5.0% | #296 |
| SciCode |