The matter fabricator creates extra matter stabilizers when productivity moduels are used. This means you get more and more of them over time, which is almost certainly not intended. This should be handled like the kovarex recipe, which doesn't create extra u-238

U can't use productivity modules in matter fabricator. What version are u using? Have u enabled beacon productivity (that could create this bug?)

This was with 1.3.0 - productivity modules are directly in the fabricator plants, "beacon productivity" setting is off.
Same thing after upgrading to 1.3.1: I can remove and re-add productivity modules without an issue, other than stabilizer multiplication.

Also, not being able to put productivity mods in them sounds like a very undesirable balancing change to me :)

In any case, fixing the recipe will fix it with beacons too, so that sound sot me like the obvious way to proceed.

Will be fixed in the next update.
For balance, the chance to lose a stabilizer during crafting was reduced.

Indeed, this was "fixed", unfortunately not by fixing the recipe but by nerfing the recipe to no longer allow productivity modules. Unfortunately, "the best way to do science" might no longer be true now.

Right, I don't understand the need for this brutal nerf - after loading my game, science production went down majorly, invalidating large parts of the optimisation I built into my plants. This change seems so unnecessary and erratic that I lost confidence in playing with this mod. It just feels like introducing arbitrary breakage.

Krastorio gave an option to turn productivity beacons on. Turning that on should let you solve the problem.

In the next update, recipes for the production of science packages will no longer require stabilizers and can use productivity modules. I hope this solves the problem.

Thanks, but out of curiosity, why not simply fix the recipe to work properly with productivity modules? Is there something special about these recipes so that marking the stabiliser as a catalyst doesn't work?

I did not know that it was possible to solve the problem by marking the stabilizer as catalyst. Thanks!