deepseek-math-7B-instruct [bfloat16] [rev: 0a5828f800a3]: benchmark results
Provider: DeepSeek. Access: Open.
Aggregate Rating 1331 ± 36, rank #8146 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 19.41 | Accuracy (%) (5-shot) | 40.6 |
| Open LLM Leaderboard v1 - MMLU | 55.97 | Accuracy (%) (5-shot) | 37.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 53.41 | Normalized accuracy (%) (25-shot) | 29.4 |
| Open LLM Leaderboard v1 - HellaSwag | 71.5 | Normalized accuracy (%) (10-shot) | 21.6 |
| Open LLM Leaderboard v1 - WinoGrande | 65.75 | Accuracy (%) (5-shot) | 19.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 40.16 | MC2 (%) (0-shot) | 13.1 |
Interactive version: theaggregate.ai/model?slug=deepseek-math-7b-instruct-bfloat16-rev-0a5828f800a3 · How It Works · Data refreshed daily, snapshot 2026-11-08.