GSM8K Chat

Grade School Math 8K adapted for chat format evaluation, featuring 8.5K high-quality linguistically diverse grade school math word problems requiring multi-step reasoning and elementary arithmetic operations.

Llama 3.1 Nemotron 70B Instruct from NVIDIA currently leads the GSM8K Chat leaderboard with a score of 0.819 across 1 evaluated AI models.

Paper
About this benchmark

What GSM8K Chat measures

GSM8K Chat is a text benchmark that evaluates large language models on math and reasoning tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.8, with the leader reaching 0.8.

Compare leaders on the best AI for math and best AI for reasoning leaderboards.

Publication

Paper
Training Verifiers to Solve Math Word Problems
Authors
Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, and 8 others
Published

Abstract

State-of-the-art language models can match human performance on many tasks, but they still struggle to robustly perform multi-step mathematical reasoning. To diagnose the failures of current models and support research, we introduce GSM8K, a dataset of 8.5K high quality linguistically diverse grade school math word problems. We find that even the largest transformer models fail to achieve high test performance, despite the conceptual simplicity of this problem distribution. To increase performance, we propose training verifiers to judge the correctness of model completions. At test time, we generate many candidate solutions and select the one ranked highest by the verifier. We demonstrate that verification significantly improves performance on GSM8K, and we provide strong empirical evidence that verification scales more effectively with increased data than a finetuning baseline.

Progress Over Time

Interactive timeline showing model performance evolution on GSM8K Chat

State-of-the-art frontier
Open
Proprietary

GSM8K Chat Leaderboard

1 models
ContextCostLicense
170B
Notice missing or incorrect data?

FAQ

Common questions about GSM8K Chat.

What is the GSM8K Chat benchmark?

Grade School Math 8K adapted for chat format evaluation, featuring 8.5K high-quality linguistically diverse grade school math word problems requiring multi-step reasoning and elementary arithmetic operations.

What is the GSM8K Chat leaderboard?

The GSM8K Chat leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, Llama 3.1 Nemotron 70B Instruct by NVIDIA leads with a score of 0.819. The average score across all models is 0.819.

What is the highest GSM8K Chat score?

The highest GSM8K Chat score is 0.819, achieved by Llama 3.1 Nemotron 70B Instruct from NVIDIA.

How many models are evaluated on GSM8K Chat?

1 models have been evaluated on the GSM8K Chat benchmark, with 0 verified results and 1 self-reported results.

Where can I find the GSM8K Chat paper?

The GSM8K Chat paper is available at https://arxiv.org/abs/2110.14168. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does GSM8K Chat cover?

GSM8K Chat is categorized under math and reasoning. The benchmark evaluates text models.

What is the best open-source model on GSM8K Chat?

Llama 3.1 Nemotron 70B Instruct by NVIDIA is the top-ranked open-source model on GSM8K Chat, with a score of 0.819 (rank #1).

How recent are the GSM8K Chat leaderboard results?

The GSM8K Chat leaderboard was last updated in May 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all math
GPQA

A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.

reasoning
216 models
MMLU-Pro

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.

math
120 models
AIME 2025

All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.

math
109 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

math
99 models
SWE-Bench Verified

A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.

reasoning
92 models
Humanity's Last Exam

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

mathmultimodal
77 models