Previous studies have associated microbial alterations with gastritis; however, problems regarding the reproducibility of biomarkers in cohorts remain poorly resolved. Patients from different regions provided excellent information, that was conducive to the exploration of the gastric microbes associated with gastric inflammation. We performed a meta-analysis of seven reliable datasets, consisting of a total of 367 fecal 16s rRNA including atrophic gastritis (AG) (174 patients) and superficial gastritis (SG) (170 patients) after quality control and discarding samples of gastric fluid, since the microbial diversity between gastric fluid and antrum was different. We also investigated the similarity in microbial composition between healthy controls and SG without Helicobacter pylori infection, but not in those with H. pylori infection. According to the linear discriminant analysis effect size and random forest analyses, Bacteroides, Weissella, Actinomyces, Atopobium, Oribacterium, Peptostreptococcus, and Rothia were biomarkers between SG and AG (AG_N) without H. pylori infection, whereas Actinomyces, Cutibacterium, Helicobacter, Novosphingobium, Pseudomonas, and Streptococcus were signatures between SG and AG (AG_P) with H. pylori infection. We also found that the development of AG_N was associated with lower vitamin B12 levels, whereas significant changes in the gastric microbiota functions of AG_P included gallate, amino acid degradation, and heme biosynthesis. In addition, non-dialysis classifiers were developed for AG_N with an area under the curve (AUC) of 0.7417 and for AG_P with an AUC of 0.8862. This study thoroughly profiled the gastric microbiota of AG and SG and provided evidence of the potential of microbial markers as noninvasive classifiers for global AG.