Design recommendations ENV 1993-1-2 was drafted in the early 1990s, based on the information available at that time. Since then, a substantial amount of research has been carried out in the field of fire modelling and steel and steel-concrete composite structures in fire. The ECCS model code on fire engineering [24] is a recent document prepared for the ECCS Technical Committee 3 Fire safety of steel structures by European fire experts. The document is to serve as the basis for converting ENV 1993- 1-2 into its full EN status, and it is for this reason that it was decided also to use the ECCS model code as a basis for the design recommendations for stainless steel. 4.1. Columns The ECCS model code recommends (for carbon steel) that the design buckling resistance Nb, f i,t,Rd of a compression member at time t is given by Eq. (2). Nb, f i,t,Rd = χ f i Aky,θ fy/γM, f i (2) where χ f i is the reduction factor for flexural buckling in the fire situation, A is the cross-sectional area (or effective area) of the member, fy is the room temperature yield strength of the material, ky,θ is the yield strength reduction factor (based on a 2% total strain limit for Class 1–3 cross-sections and 0.2% plastic strain limit [4] forClass 4 cross-sections) at temperature θ, and γM, f i is the partial safety factor for the fire situation (recommended to be taken as 1.0). Herein, it is proposed to extend the recommendations of the ECCS model code to stainless steel columns, by adopting the above formulations, but with stainless steel material properties, as reported in Tables 1–5 of this paper. A comparison between the proposed fire buckling resistance (at the critical temperature reached in the actual fire test) of each of the six columns, and the applied test loads is given in Table 13. In the comparisons, the measured geometric and material properties are employed and all partial factors are set equal to unity. The comparison shows, for the five hollow section members, a mean applied test load divided by proposed fire buckling resistance of 1.37, implying that the above formulations can be safely employed for stainless steel, but that there may be scope for further improvements to the design expressions. In particular, the Class 4 200 × 200 × 4 RHS, for which an effective area and the elevated temperature 0.2% proof strength [4] (in place of the strength at 2% total strain which is used for Class 1–3 cross-sections), performed significantly better than predicted by the proposed design expressions. However, given the scarcity of the test data and for consistency with carbon steel, it is currently recommended that the above formulations be adopted unmodified.