Abstract Background Timely and accurate monitoring of renal dysfunction is critical for patients with chronic kidney disease (CKD). Most equations currently used for estimating glomerular filtration rate (GFR) incorporate serum levels of creatinine and/or cystatin C. Here, we evaluate a recently introduced method for estimating GFR that uses nuclear magnetic resonance spectroscopy (NMR) to measure creatinine, myo-inositol, and valine in combination with cystatin C measured by immunoassay, and patient’s age and sex (eGFRNMR). Methods Serum creatinine, valine, and myo-inositol measurements were performed simultaneously by an NMR assay on a Bruker Avance III HD 600 MHz NMR instrument (Bruker Corporation, MA). Cystatin C testing was performed using an immunoturbidimetric assay (Diazyme Laboratories, CA) on a cobas 8000 analyzer (Roche Diagnostics, IN). Within-run and between-day precision studies were performed across 5 days. Method comparison was carried out by testing 86 split serum samples from adult individuals against the numares AG GFR (NMR) assay. Of the 86 samples, 20 were spiked with various concentrations of creatinine, myo-inositol, and valine to reach a wide range of GFR results. An additional method comparison study was performed against Mayo Clinic Laboratories (Rochester, MN) on 59 serum samples from adult patients that had already been tested using the same assay. Measured GFR (mGFR) was also performed on these patients by Mayo Clinic Laboratories using urinary iothalamate clearance. Results Within-run and between-day coefficients of variation (%CV) for eGFRNMR did not exceed 6%. The eGFRNMR measurement range was confirmed at 10 to 157 mL/min/1.73 m2 based on verified analytical measurement ranges of creatinine (0.28–9.8 mg/dL), myo-inositol (39–430 µmol/L), and valine (30–1250 µmol/L). Method comparison studies against both numares AG and Mayo Clinic Laboratories demonstrated excellent concordance (r2 = 0.95 and r2 = 0.97, respectively) and minimal bias (4.1% and −0.3%, respectively) with results ranging from 21 to 157 mL/min/1.73 m2. When the 59 patients who had their GFR measured were staged by their mGFR results according to the CKD classification ranges of <15, 15–29, 30–44, 45–59, 60–89, and ≥90 mL/min/1.73 m2, eGFRNMR showed higher classification concordance with mGFR (69%) than the CKD-EPI 2021 creatinine-cystatin C equation (63%). Conclusion Our evaluation confirms the excellent analytical and clinical performance of the recently introduced eGFRNMR assay, supporting its integration into clinical practice for improved estimation of GFR.