Overall, I was very impressed with the written manuscript and the scientific rigor of the analysis. The introduction and methods clearly describe the complex ML, ensemble, and processed based ET estimates as well as training and validation analysis. The results showed high accuracy, and the figures were well developed and easily visualized. I have minor comments below:

Line 100: Precipitation was not used as an input covariate. Please provide an explanation for why this was not included.

Line 110: You do a great job outlining the models available with Autogluon. In line 110, you end the list with “etc.” – leading me to believe that there are even more algorithms available. In line 111, can you please list the specific ML algorithms you used?

Line 110: You say “Autogluon can combine them” – can you provide more specifics on this (perhaps just changing the phrasing) – did Autogluon combine them in your research OR autogluon can combine them – but you did not in your research. I think simply stating “Autogluon combined all the algorithms mentioned above” would suffice.

Figure 1: This is a great workflow figure – and helps visualize the process.

Line 164: You say you used 6 well known ET products – can you please cite them after the sentence in line 164? Or perhaps say “refer to section 3.3”

Figure 5: Can you include the sites and land cover types of the three lowest RMSE in Line 350 paragraph. It would provide additional detail than “the majority of land covers”. I think even reference table 3 in the paragraph of Line 350 would be helpful.

Table 2: This seems repetitive and unnecessary. It is unclear how it is different than Figure 4.

The overall importance of the paper is not clear. Are the global ET estimates available for public use – if so, is a link available to the dataset and what is the spatial and temporal resolution? If the estimates are available – I recommend making it clear and provide specific details and explanations about the use of the data.

On the other hand, is the paper simply a methodologic paper meant to explain a novel method for estimating global ET –although you explain the research, it would be hard for another researcher to reproduce for local or global ET estimates. If so, I recommend making it clear that the data are meant to remain proprietary, but the manuscript simply provides novel methodology.

This study uses a Classifier-Guided Ensemble model to integrate process-based algorithm, ML-based model, and Hybrid model to identify the best model for a given case and subsequently estimate evapotranspiration (ET). The results demonstrate that the selected dominant model outperforms the individual constituent models. And the contribution of input parameters is also analyzed, which identifies the key drivers across all 3 model types. Overall, this study offers a tool to select the better model for a specific location among different models. Here, I have some comments and suggestions for the authors to consider in improving the manuscript, as well as some questions:

1. Line 142: The manuscript mentioned “ET as the sum of daytime and nighttime components”, but the calculation methods for these two components are not clearly described. Please provide more details on how daytime and nighttime ET are estimated and combined.
2. Lines 163-164: How did you extract the relatively reliable data from global ET products?
3. Lines 168-169: The manuscript mentioned “data from other pixels were excluded to reduce uncertainty”, what is the meaning of other pixels? Do you mean the pixels from other finer products but within the same pixel already settled?
4. Lines 171-173: You get the dominant model for both time and spatial scales. If for the same spatial, the dominant model varies frequently with time, what is your strategy for forecasting? Do you select the dominant model for a specific location with more better-results?
5. Line 233: The variables from ERA5-land product were used for training. Were the input variables for the other five models derived from the same data source? If not, were inter-dataset comparisons conducted? If differences among variables are substantial, the training process may introduce model bias.
6. Figure 5: The colors of Classifier-Guided and Hybrid model are very similar in the axis plots. Using more distinct colors would improve the clarity and comparability of model performance.
7. Lines 557-558: The sentence appears to be incomplete and should be revised for clarity.