xglm-7.5B: benchmark results
Meta's XGLM-7.5B, a multilingual autoregressive language model trained on a balanced 500-billion-sub-token corpus across diverse languages, introduced in Meta's few-shot multilingual language model research. Provider: Meta. Released 2021-12-28. Access: Open.
Aggregate Rating 1187 ± 12, rank #1577 of 1635 rated models, from 89 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - High School Statistics | 42.13 | Accuracy (%) | 62.8 |
| MMLU-by-task - Formal Logic | 31.75 | Accuracy (%) | 56 |
| MMLU-by-task - Professional Medicine | 45.22 | Accuracy (%) | 48.2 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| MMLU-by-task - High School Physics | 29.8 | Accuracy (%) | 46.5 |
| MMLU-by-task - High School Chemistry | 31.03 | Accuracy (%) | 43.8 |
| MMLU-by-task - High School Macroeconomics | 35.64 | Accuracy (%) | 36.4 |
| MMLU-by-task - Elementary Mathematics | 26.72 | Accuracy (%) | 36.2 |
| MMLU-by-task - High School Microeconomics | 33.19 | Accuracy (%) | 35.6 |
| MMLU-by-task - High School Biology | 33.23 | Accuracy (%) | 35.5 |
| MMLU-by-task - Abstract Algebra | 27 | Accuracy (%) | 34 |
| MMLU-by-task - High School Geography | 35.86 | Accuracy (%) | 33.8 |
Interactive version: theaggregate.ai/model?slug=xglm-7-5b · How It Works · Data refreshed daily, snapshot 2026-10-15.