Codesota · Models · Apertus-70B-Instruct7 results · 7 benchmarks
Model card
Apertus-70B-Instruct.
unknown
§ 02 · Benchmarks
Every benchmark Apertus-70B-Instruct has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | DROP | Natural Language Processing · Question Answering | f1 | 50.8% | #4 | — | source ↗ |
| 02 | MBPP+ | Computer Code · Code Generation | pass-1 | 47.0% | #7 | — | source ↗ |
| 03 | BIG-Bench Hard | Reasoning · Multi-step Reasoning | accuracy | 64.2% | #9 | — | source ↗ |
| 04 | HellaSwag | Reasoning · Commonsense Reasoning | accuracy | 78.1% | #11 | — | source ↗ |
| 05 | GSM8K | Reasoning · Mathematical Reasoning | accuracy | 77.6% | #39 | — | source ↗ |
| 06 | MATH | Reasoning · Mathematical Reasoning | accuracy | 30.8% | #43 | — | source ↗ |
| 07 | MMLU | Reasoning · Commonsense Reasoning | accuracy | 69.6% | #54 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where Apertus-70B-Instruct actually performs.
§ 04 · Papers
1 paper with results for Apertus-70B-Instruct.
- 2025-09-17· 7 results
Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
§ 06 · Sources & freshness
Where these numbers come from.
pwc-dump
7
results
0 of 7 rows marked verified.