Text-based nutrient estimation by large language models: a comparative evaluation against the Chinese Food Composition Table.
- 1Gansu Provincial Maternal and Child Health Care Hospital (Gansu Provincial Central Hospital), Lanzhou, Gansu Province, China.
- 2Gansu University of Chinese Medicine, Lanzhou, Gansu Province, China.
Abstract
Background
Large language models (LLMs) increasingly provide nutrient information, but their accuracy for Chinese foods and their consistency across repeated queries are not well established.Objective
To compare nutrient estimates from nine general-purpose LLMs with the Chinese Food Composition Table (CFCT) and assess within-model reproducibility.Methods
For 299 commonly consumed Chinese foods, each model estimated energy and 14 nutrients per 100 g in three independent runs (8,073 food-level outputs). Model means were compared with CFCT values using absolute-agreement intraclass correlation coefficients (ICCs), bias, mean absolute error (MAE), root mean squared error, percentage error, and Bland-Altman limits of agreement. Repeated-run ICCs and coefficients of variation assessed reproducibility.Results
Agreement was excellent for energy, fat, carbohydrate, sodium, and vitamin C across all nine models, and good to excellent for protein. Performance was less consistent for vitamin A, zinc, iron, and potassium. Repeated-run stability was high for energy and macronutrients but lowest for vitamin A. GLM had the most favorable combined accuracy-stability profile, followed by Minimax and Qwen. Even when aggregate agreement was high, individual food estimates sometimes showed substantial error.Conclusion
General-purpose LLMs can approximate energy and macronutrient values for Chinese foods under controlled text-based conditions, but micronutrient estimates remain less dependable. Values intended for clinical or other high-stakes applications should be verified using validated food-composition resources and, where appropriate, qualified nutrition professionals.
Related
- Nutrient Calculation Accuracy, Expert-Rated Clinical Appropriateness and Safety Indicators, and Reproducibility of Large Language Model-Generated Diet Plans for Polyendocrine Metabolic Ovarian Syndrome: A Real-Case-Based Evaluation
- Comparative Evaluation of Large Language Model Interfaces in Third Molar Surgery Complication Scenarios: Response Quality, Clinical Content, Potential Clinical Risk, Readability, and Externally Observable Response Latency—A Cross-Sectional Comparative Benchmark Study
- Comparative Evaluation of Large Language Models in Responding to Common Misconceptions Regarding Periodontal Disease and Oral Hygiene
- Citation Fabrication and Accuracy in AI-Generated Psychiatric Research: A Comparative Evaluation of Four Contemporary Large Language Models
- A Multidimensional Evaluation of Large Language Model Responses to the 2024 ESC Hypertension Guidelines: A Comparative Study
BibTeX
@article{cao2026textbased,
title = {{Text-based nutrient estimation by large language models: a comparative evaluation against the Chinese Food Composition Table.}},
author = {Yongwen Cao and Zhendong Liu and Ning Wang and Min Chen and Hongyan Wang},
journal = {Frontiers in nutrition},
year = {2026},
month = oct,
doi = {10.3389/fnut.2026.1952757},
url = {https://europepmc.org/article/MED/42898137},
note = {CC BY},
}
RIS
TY - JOUR
TI - Text-based nutrient estimation by large language models: a comparative evaluation against the Chinese Food Composition Table.
AU - Yongwen Cao
AU - Zhendong Liu
AU - Ning Wang
AU - Min Chen
AU - Hongyan Wang
PY - 2026
DA - 2026/10/05
JO - Frontiers in nutrition
DO - 10.3389/fnut.2026.1952757
UR - https://europepmc.org/article/MED/42898137
AB - Background Large language models (LLMs) increasingly provide nutrient information, but their accuracy for Chinese foods and their consistency across repeated queries are not well established.Objective To compare nutrient estimates from nine general-purpose LLMs with the Chinese Food Composition Table (CFCT) and assess within-model reproducibility.Methods For 299 commonly consumed Chinese foods, each model estimated energy and 14 nutrients per 100 g in three independent runs (8,073 food-level outputs). Model means were compared with CFCT values using absolute-agreement intraclass correlation coefficients (ICCs), bias, mean absolute error (MAE), root mean squared error, percentage error, and Bland-Altman limits of agreement. Repeated-run ICCs and coefficients of variation assessed reproducibility.Results Agreement was excellent for energy, fat, carbohydrate, sodium, and vitamin C across all nine models, and good to excellent for protein. Performance was less consistent for vitamin A, zinc, iron, and potassium. Repeated-run stability was high for energy and macronutrients but lowest for vitamin A. GLM had the most favorable combined accuracy-stability profile, followed by Minimax and Qwen. Even when aggregate agreement was high, individual food estimates sometimes showed substantial error.Conclusion General-purpose LLMs can approximate energy and macronutrient values for Chinese foods under controlled text-based conditions, but micronutrient estimates remain less dependable. Values intended for clinical or other high-stakes applications should be verified using validated food-composition resources and, where appropriate, qualified nutrition professionals.
SN - 2296-861X
ER -