internlm2-7B: benchmark results
Shanghai AI Lab's 7B InternLM2 base model (January 2024), notable for near-perfect needle-in-a-haystack retrieval at 200K-token context. Provider: Shanghai AI Lab. Released 2024-01-17. Access: Open.
Aggregate Rating 1409 ± 17, rank #1192 of 1656 rated models, from 60 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Chinese LLM - WinoGrande | 70.01 | Accuracy (%) | 91.6 |
| Open LLM Leaderboard v1 - WinoGrande | 83.82 | Accuracy (%) (5-shot) | 90.8 |
| Open LLM Leaderboard v1 - MMLU | 65.24 | Accuracy (%) (5-shot) | 83.3 |
| Open LLM Leaderboard - GPQA | 33.67 | Score | 83.2 |
| Open LLM Leaderboard - BBH | 58.25 | Score | 79.4 |
| Open Chinese LLM - HellaSwag | 65.44 | Accuracy (%) | 79 |
| Open LLM Leaderboard - MuSR | 44 | Score | 78.1 |
| Open LLM Leaderboard v1 - GSM8K | 63 | Accuracy (%) (5-shot) | 76.3 |
| Open Chinese LLM Leaderboard | 59.83 | Official Nonzero-Task Mean (%) | 67.5 |
| Open Chinese LLM - GSM8K | 49.66 | Accuracy (%) | 66 |
| Open Chinese LLM - ARC Challenge | 54.52 | Accuracy (%) | 65.8 |
| Open Chinese LLM - CMMLU | 57.52 | Accuracy (%) | 64.5 |
Interactive version: theaggregate.ai/model?slug=internlm2-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.