This paper describes the general set-up and equations involved in a new zonally-averaged Energy Balance Model (EBM) designed for paleo-climatic studies. The paper is well-organized and well-written. The results are clearly presented and discussed in a rather balanced way, showing both the successes and the limitations of this new model. Overall, I only have a few minor suggestions listed (unsorted) below and I consider that this manuscript could be published with only a few  corrections.

- equation (1): Ls should be replaced by Lf (Latent heat of fusion, described later on).

- the description of the Hadley cell circulation is a bit difficult to follow with the use of many intermediate notations for heat fluxes. Maybe a final summary of FT and FQ as functions of Ftotal , FT_eddy and FQ_eddy would be useful (equations 18a, 19a).

- there is apparently no vegetation on land and the surface albedo is controlled only by snow… I am wondering to what extent this explains some biaises of the model, or if this is negligible when compared to other factors (cloud cover, fresh snow versus ice albedo, …). More importantly, it is not clearly explained in the manuscript how land evaporation is computed: there is a “surface water availability” parameter in equation (7), but no information is given on what it actually means. A bit more discussion on land cover (or lack of) would be appreciated.

- Table 1 lists only a small set of the parameters used in the model. It would be useful to have a more extended list…Examples:

line 177: Ta’ is Ta corrected with a lapse rate of -6.5K/km. This information is useless if we don’t know the height at which Ta’ is evaluated.

Line 309: constant sea ice thickness…

- cloud cover is taken from NorESM and is shown of Fig.1 along some other fixed parameters (ocean circulation, Hadley cells). There is some discussion in the paper of the impact of ocean circulation… but little discussion on the Hadley cell parameters, and none on the clouds. This should certainly be addressed in a revised version of the paper.

- line 174: fsf is the fraction of precipitation that falls as snow. Usually, this is understood as a statistical, or possibly as a time fraction. But later on (line 233) this is used as a geographic fraction, which might be something quite different… Is this really justifiable?

- the results in terms of snow cover (Figure5, lines 385) are a bit disappointing. Concerning a future application of this model to the question of ice ages (like a coupling with an ice-sheet model), this might be a severe limitation. Some discussion on this point would be useful… Can really the model be used for such a purpose?

- almost all model outputs are compared to GCM outputs, except for the meridional heat fluxes on Figure 9c and 9d. This is a bit surprising and, if possible, it should be corrected.

This paper uses a new energy and moisture balance model, ZEMBA, to simulate a range of climate states: the pre-industrial, 2xCO2, ±2% top-of-atmosphere solar radiation, and the Last Glacial Maximum (LGM). In the atmosphere, heat is transported based on meridional gradients in moist-static-energy, and in the ocean, heat is transported according to prescribed vertical ocean velocities. Importantly, ZEMBA also includes a hydrologic cycle. The manuscript is well-written, the novelty of the approach is clearly demonstrated, and the presented model applications are appropriate. However, I have a few minor suggestions prior to publication.

The model does have some limitations related to its parameterizations, including using a single cloud cover fraction from pre-industrial simulations. One test of the importance of this feature could be testing a range of possible cloud cover fractions (from 2xCO2 experiments, for instance) in ZEMBA. Brief mention of which parameters in Table 1 have (1) high uncertainty and (2) significant impact on the modeled climate would be helpful, as would references for the chosen parameter values. How are the parameters tuned, and are their values within the accepted range of uncertainty? For example, the chosen diffusion coefficient Da is slightly higher than estimates from GCMs (~1.05 x 10^6, e.g., Ge et al., 2023; “The sensitivity of climate and climate change to the efficiency of atmospheric heat transport”).

Additionally, some clarification on the impact and inclusion of the seasonal cycle would be useful. Is the seasonal cycle being solely driven by insolation changes, or do other parameters change as well? Are simulated climate significantly different if annual-mean insolation values are used? In Siler et al. (2018), only annual-mean precipitation and evaporation patterns were modeled (not seasonal variations); given ZEMBA’s underestimation of snow coverage over land (Fig. 5), is the inclusion of seasonal hydrology reasonable?

Line comments:

Line 95 – “its” not “it’s”

Line 555 – there is a model available that couples a carbon cycle to an EBM with a hydrologic cycle, and simulates ice sheet growth and decay (Kukla et al; “All aboard! Earth system investigations with the CH2O-CHOO TRAIN v1.0”). A reference should be included, and would recommend rephrasing “There are comparatively few EBMs which incorporate a hydrological cycle (Jentsch, 1991) and none – to our knowledge – used for studies of glacial-interglacial cycles.”