internlm2-1.8B: benchmark results
Shanghai AI Lab's InternLM2-1.8B, the foundation base model in a three-version 1.8B family (base, Chat-SFT and Chat), supporting contexts up to 200,000 characters. Provider: Shanghai AI Lab. Released 2024-01-31. Access: Open.
Aggregate Rating 1237 ± 37, rank #1552 of 1654 rated models, from 38 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open PL LLM - PSC (multiple choice, 5-shot) | 44.6 | Binary F1 (%) | 30.7 |
| Open LLM Leaderboard - MuSR | 38.13 | Score | 30 |
| Open Portuguese LLM - FaQuAD NLI | 46.68 | Macro F1 (%) | 29 |
| Open Portuguese LLM - BLUEX | 31.71 | Accuracy (%) | 27.8 |
| Open PL LLM - PSC (generative, 5-shot) | 76.07 | Binary F1 (%) | 26.8 |
| Open Portuguese LLM - OAB Exams | 30.75 | Accuracy (%) | 26.8 |
| Open Portuguese LLM - ASSIN2 RTE | 72.34 | Macro F1 (%) | 26.1 |
| Open Portuguese LLM - ENEM | 29.11 | Accuracy (%) | 23.7 |
| Open LLM Leaderboard - BBH | 38.8 | Score | 23 |
| Open PL LLM - EQ-Bench | 2.83 | EQ-Bench Score | 20.8 |
| Open PL LLM - DYK (generative, 5-shot) | 28.97 | Binary F1 (%) | 19.5 |
| Open PL LLM Leaderboard | 3.15 | Average Score (%) | 18.9 |
Interactive version: theaggregate.ai/model?slug=internlm2-1-8b · How It Works · Data refreshed daily, snapshot 2026-10-24.