The metabolisable protein (PDI) supplies were revised in the new INRA feeding system (2013). Consequently, the digestible amino acid (AA) supply in the intestine (AADI, in % of PDI) for cattle was to revise since it was developed in 1998. A meta-analysis was performed to estimate the AADI using a close approach to the past method. The new database is larger than the past one (30 publications and 133 diets from 1967 to 1992) since it includes 64 publications (from 1967 to 2013) on cattle and 243 dietary treatments mainly focussed on nitrogen supply (amount or quality). Each dietary AA supply in the duodenum (g/kg DM) was calculated using a linear combination of the rumen undegraded (RUP), the microbial and the endogenous protein flows in the intestine (PI) with the individual feed AA profiles and the two fixed microbial and endogenous profiles, respectively. It was then expressed in % of total AA (i.e. 16 AA). Each relationship between the measured and the calculated AA supply was analysed using a linear model of covariance with a fixed publication effect. The new vs the past models were for the 4 AA recommended in the INRA system: 0.80 + 0.90 × X (RMSE=0.29) vs 1.90 + 0.76 × X (RMSE=0.38) for Lys; 1.03 + 0.45 × X (RMSE=0.17) vs 0.32 + 0.73 × X (RMSE=0.26) for Met; 0.57 +0.79 × X (RMSE=0.24) vs 0.93 + 0.58 × X (RMSE=0.22) for His; 1.78 + 0.86 × X (RMSE=0.32) vs 2.55 + 0.76 × X (RMSE=0.45) for Leu, respectively. The new INRA PI estimations led to lower biases between the measured and the calculated AA supplies compared to the past models since lower RMSE were observed for 15 AA models in addition to a slope closer to one and an intercept closer to zero for 12 AA models. Further analyses will describe the interfering factors influencing the relationships observed for Met, Phe, Arg, Gly, and the RMSE for His. It will be possible to improve the AADI calculations by using these new models to estimate the AA in RUP and microbial PI and the digestibilities of RUP and microbial PI.