OLMo-7B-Instruct-hf: benchmark results
Allen AI's OLMo 7B Instruct, adapted from the Dolma-trained OLMo base model via the Tulu SFT mixture and a cleaned UltraFeedback dataset; superseded by newer OLMo releases. Provider: Allen AI. Released 2024-06-04. Access: Open.
Aggregate Rating 1227 ± 29, rank #1565 of 1656 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Enkrypt AI - Bias Risk | 83.2 | Risk Score | 52.7 |
| Open LLM Leaderboard - IFEval | 34.73 | Score | 33.1 |
| Open LLM Leaderboard - MuSR | 37.65 | Score | 27.6 |
| Open LLM Leaderboard - GPQA | 27.1 | Score | 25.8 |
| Open LLM Leaderboard - BBH | 37.06 | Score | 21.4 |
| Open LLM Leaderboard - MMLU-Pro | 17.85 | Score | 19.3 |
| BenCzechMark - NER | 15.58 | Duel Win Score (%) | 16.9 |
| Open LLM Leaderboard - MATH Level 5 | 1.36 | Score | 9.9 |
| BenCzechMark - Reading Comprehension | 6.93 | Duel Win Score (%) | 9.7 |
| BenCzechMark | 6.57 | Average Duel Win Score (%) | 7.8 |
| BenCzechMark - NLI | 10.63 | Duel Win Score (%) | 7.8 |
| BenCzechMark - Sentiment | 7.47 | Duel Win Score (%) | 6.5 |
Interactive version: theaggregate.ai/model?slug=olmo-7b-instruct-hf · How It Works · Data refreshed daily, snapshot 2026-10-20.