As a worldwide public health issue, cancer‐induced cachexia can result in decreasing physical function and survival rate. However, the therapeutic effects of conventional approaches, including pharmacotherapy, exercise and nutritional intervention, are far from satisfactory. Herbal medicines (HMs), especially Traditional Chinese Medicine (TCM), are reported to effectively treat cachexia for centuries. The inclusion criteria of all participants in this study pointed to the diagnosis of cachexia, the trial group used herbal medicine (HM) in complementary and alternative medicine, etc. Twelve databases, including EMbase, PubMed, Web of science, Cochrane CENTRAL, CINAHL, CINAHLPlus, PsycINFO, AMED, China Biology Medicine disc (CBM), China National Knowledge Infrastructure (CNKI), Wanfang and Chongqing VIP (CQVIP) were retrieved from inception to March 28, 2022. We conducted the meta‐analysis utilizing RevMan 5.3. A trial sequential analysis (TSA) was conducted to assess the adequacy of the sample size for the outcomes. We have registered the protocol and the registration number was CRD42022336446. A total of 66 studies were included, containing 3654 patients diagnosed with cancer cachexia, of which 1833 patients were assigned to the trial group and 1821 patients were treated in the control group. Outcomes cover the primary indicator KPS (RR = 1.84, 95%CI = [1.61, 2.09], p < 0.00001), and other outcomes including adverse events rate (RR = 0.37, 95%CI = [0.23, 0.58], p < 0.0001), albumin (MD = 2.14, 95%CI = [1.56, 2.71], p < 0.00001), haemoglobin (MD = 4.88, 95%CI = [3.26, 6.50], p < 0.00001), TCM syndrome effect (MD = 1.47, 95%CI = [1.31, 1.65], p < 0.00001), effect of weight (RR = 1.62, 95%CI = [1.34, 1.95], p < 0.00001), effect of appetite (RR = 1.23, 95%CI = [1.13, 1.34], p < 0.00001), FAACT (RR = 7.81, 95%CI = [6.12, 9.50], p < 0.00001), PG‐SGA (MD = −2.16, 95%CI = [−2.65, −1.67], p < 0.00001) and QOL (MD = 5.76, 95%CI = [4.04, 7.48], p < 0.00001), suggesting that HMs or HMs combined with conventional treatment have an ameliorating effect on cachexia in each respect. Subgroup analysis showed that the five HMs with the best effect on improving KPS and their optimal doses were Coicis Semen (Yiyiren) in 10 g group, Citri Reticulatae Pericarpium (Chenpi) in 15 g group, Dioscoreae Rhizoma (Shanyao) in 10 g group, Ophiopogonis Radix (Maidong) in 10 g group and Ginseng Radix Et Rhizoma (Renshen) in 20 g group. In addition, there were HM combinations of levels 2–6. Egger's test showed publication bias for five outcomes. HMs have a significant effect on improving cancer cachexia on FAACT, TCM syndrome, KPS, QOL, appetite, nutritional status (evaluated by PG‐SGA scale), weight, levels of albumin and haemoglobin. And the Adverse events rate is less than that of Western Medicine. The herbs with the best curative effect and their optimal dose were Dioscoreae R. (10 g), Citri R.P. (15 g), Coicis S. (10 g), Ophiopogonis R. (10 g) and Ginseng R.E.R. (20 g). Due to the quality of included studies is not high, further high‐quality studies are needed to firmly establish the clinical efficacy of HM.