I am doing a model using ModelBuilder in ArcMap, it is a really big model and for some reason it get corrupted every time i get certain number of tools.
there are no listed limits that I can find.
So, some questions that might help
my model works perfectly and the script works well also..the only issue is not showing in my gdb and ofcourse not showing in the TOC.
Make sure OutputFC is not set to intermediate.
Also: Why not create a small Python script that can import your toolbox with arcpy.ImportToolbox() and call your model tool from Python.
You didn't say how big your model is or post a model, but my experience is that really large models get unstable. ModelBuilder has to validate the entire model when you open it, making sure are parameters are valid and datasets are there. If a model gets too large this process can bog down quite a bit.
I agree with Dan that you should break the model into pieces and have a driver model that calls each piece in turn. Smaller submodels much easier to edit, debug and run, and they will not "freeze" while you wait for dozens of tools and parameters to validate.
Creating smaller pieces also has has the added benefit that the pieces are much more re-usable than a big behemoth model. This is also true for Python scripting!
My model is not big and my model runs perfectly with results but for me to filter selections, made me to export to python and use tool validator to filter.
We had over 150 tools in the model and in certain number of tools it just crashed and after that I was not able to edit the model, so I had to do dan's suggestion i broke it down into submodels and after that it worked, I am creating a phyton script to avoid this.
I would say once a model is exceeding a dozen or more tools I would be seriously thinking about porting it all over to python. A model with 150 tools has some serious logic to it that would be very difficult to maintain, much easier to do it in python.
Creating a workflow in modelbuilder is easy, quick and very convenient but I think people get trapped (including myself) in that scenario of having invested too much effort in it to abandon it for python.
My recommendation is that if you are working up some new cool workflow to do something fancy and you get to about 20 tools and you "can't see the light at the end of the tunnel" then stop! Move your existing logic into python code and then continue.
Well that's what my 16 years of using modelbuilder has taught me!
Retrieving data ...