MGSM
A human-translated 250-problem GSM8K subset for comparing grade-school mathematical reasoning across ten languages.
The small, static translations are broadly supported by modern multilingual models, while large language gaps and prompt sensitivity still make MGSM useful diagnostically; newer perturbation-based successors test robustness more directly.
Performance Timeline
Longitudinal progression of model scores against human baselines.Performance & Historical Trajectory
Empirical score progression across model release dates and evaluation rounds.
| Model | Score | Date | Source Type | Provenance |
|---|---|---|---|---|
| PaLM 540B (translate-to-English CoT) | 74.7% | 2023-12-06 | vendor-reported | Source ↗ |
Human Baseline & Difficulty Horizon
Calibrated human reference points, specialist benchmarks, and ceiling thresholds.Multilingual grade-school math problem solve rate across 10 languages accounting for translation nuances.
Metric & Scoring Methodology
Verification protocols, aggregation formulas, and specialized metric variants.macro-average exact-answer accuracy across languages (%)Dataset & Compute Cost
Evaluation volume, public availability, API pricing, and local hardware requirements.$5 – $20 USD for full benchmark evaluation run on frontier APIs.
1x NVIDIA RTX 4090 (24GB) or A100 (40GB/80GB) via vLLM / SGLang
How to Run & Reproduce
Standardized evaluation protocols, CLI commands, and reproducible runner templates.lm_eval --model hf --model_args pretrained=<model_path> --tasks mgsm --batch_size autoopencompass --datasets mgsm --models <model_config># Standard API Evaluation Loop
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": prompt}],
temperature=0.0,
)When publishing results for MGSM, always report the exact prompt template, few-shot exemplar ordering, sampling temperature (temperature=0), maximum reasoning budget tokens, and the precise timestamped model snapshot ID.
Contamination & Memorization Analysis
Audit of pretraining exposure risks, memorization vectors, and refresh policies.Static fixed snapshot
Public on web / HuggingFace