Pure silky pumpkin custard — that's what these mini crustless pumpkin pies deliver at 5.1g net carbs each. I skip the crust and the spice on purpose, and they set up beautifully overnight in the fridge.
I get asked about this recipe more than almost any other during fall, and the question is always the same: where are the spices? I left them out on purpose. Every teaspoon of pumpkin pie spice blend adds carbs that I'd rather spend elsewhere, and after making these for years, I actually prefer the clean pumpkin flavor without cinnamon and nutmeg competing for attention. If you want the spice (and I get it, some people do), half a teaspoon of pumpkin pie spice stirred into the batter works perfectly. Reader Nate tried exactly that and said it brought them closer to traditional pumpkin pie.
The base is simple: eggs, heavy cream, canned pumpkin puree, and Swerve confectioners. Five ingredients, one bowl, no crust to fuss with. I pour the batter into a 12-count muffin tin and bake at 375 for 40 to 50 minutes. No need for the two-temperature method some recipes call for (high heat first, then lower). I've tested both approaches and a steady 375 gives me the same set with less fiddling.
Here's the part most people don't expect: these taste completely different after a night in the fridge. Fresh from the oven, they're soft and pudding-like, almost too delicate to pick up. After eight hours in the refrigerator, the texture firms into something much closer to actual pie filling. Sliceable, dense, cool. I honestly prefer them cold the next day, and so does my family. If you're bringing these to a gathering, make them the night before.
Two readers (Nate and Beth) independently told me that switching to a silicone muffin pan changed everything for them. No cooking spray needed, and the pies release clean after about 10 minutes in the fridge. I started using silicone after reading their comments and they were right. If you're fighting sticking issues with a metal pan, that's the fix.
For the topping, I whip heavy cream with a little vanilla and Swerve until it holds soft peaks. A dollop on each chilled pie is all you need. If you're looking for more pumpkin recipes to round out the season, my keto pumpkin pie has a proper crust, or try the keto pumpkin roll for something more dramatic. The keto pumpkin pie for one is great when I don't want a full batch. And if pumpkin isn't your thing, my keto coconut cream pie uses a similar custard approach with completely different flavor.
Everything for the filling and the whipped topping comes from a regular grocery store. The spice in the middle of the photo is optional and is not on the ingredient list.
Prep takes 10 minutes, and since they are mini and crustless, they bake in less time than a full size pumpkin pie.
How to Make Mini Crustless Pumpkin Pies
I mix everything in one bowl, pour it into a muffin tin, and bake. That's genuinely all there is to it. But a few details make the difference between pies that stick and crack versus ones that release clean and look great.
Use a silicone muffin pan if you have one. Two of my readers tested this independently and both reported clean release without any cooking spray. If you're using metal, grease it well and plan on chilling the pies for at least 10 minutes before unmolding. I fill each cavity about three-quarters full to leave room for a slight rise during baking.
The toothpick test works here, but don't wait for bone-dry. A tiny bit of moisture on the toothpick is fine because these continue setting as they cool. Overbaking dries them out and you'll see cracks across the top. I pull mine at 40 minutes and check every 3 to 4 minutes after that.
Once they're baked, let them cool in the pan completely before refrigerating. The overnight fridge step isn't optional if you want that firm, sliceable custard texture. If you love crustless desserts but sometimes want a proper base, my keto pie crust pairs well with this same filling.
Preheat oven to 375 degrees and spray cooking oil into a 12 count muffin pan.
2
Mix ingredients
In a large bowl, mix all the ingredients until smooth and combined.
3
Distribute & bake
Evenly distribute the pumpkin batter between the 12 cavities. Bake at 375 degrees for 40-50 minutes or until an inserted toothpick comes out clean.
Nutrition
Per serving: 1 slice
Calories
174
Fat
17.3g
Protein
2.1g
Total carbs
6.9g
Net carbs
5.1g
Nutrition is an estimate calculated from the ingredients listed. Brands vary, so check labels if you track closely.
Frequently Asked Questions
Why doesn't this recipe have pumpkin pie spice?
I left it out on purpose. Every teaspoon of pumpkin pie spice blend adds carbs, and after years of making these, I actually prefer the clean pumpkin flavor on its own. The eggs, cream, and pumpkin do the work without needing cinnamon and nutmeg to carry the taste. That said, if you want the traditional spice flavor, half a teaspoon stirred into the batter is all you need.
Should I use a silicone or metal muffin pan?
I switched to silicone after two of my readers (Nate and Beth) both told me their pies released perfectly without any cooking spray. They were right. Silicone lets you skip the greasing entirely, and the pies pop out clean after about 10 minutes in the fridge. If you only have a metal pan, grease it generously and chill before unmolding.
Can I make these dairy-free?
I've tested full-fat coconut milk in place of the heavy cream and it works well. The flavor shifts slightly toward coconut, but the texture stays creamy and the pies set up the same way overnight. For the whipped topping, coconut cream (the thick layer from a chilled can) whips up nicely with a little sweetener.
What's the texture difference fresh versus refrigerated overnight?
Fresh from the oven, these are soft and pudding-like. Almost too delicate to pick up without a spoon. After a night in the fridge, the texture completely changes. They firm into a dense, sliceable custard that's much closer to traditional pie filling. I prefer them cold, and I think most people will too once they try both.
Can I make this as a full-size pie instead of muffin tin?
I've poured this same batter into a 9-inch pie dish and it works. You'll need to extend the bake time to about 45 to 55 minutes since the deeper fill takes longer to set. The toothpick test still applies. I like the muffin tin version for portion control, but a full-size crustless pie is great if you're serving a crowd. Try it alongside my keto apple pie for a two-pie spread.
How do I store leftover mini pumpkin pies?
I keep mine in an airtight container in the fridge for up to 5 days, and they taste better on day 2 and 3 than fresh from the oven. You can freeze them too. I lay them in a single layer on a sheet pan to freeze solid, then transfer to a freezer bag. Thaw in the fridge overnight before serving.
Can I use a different sweetener instead of Swerve?
I've made these with allulose and with monk fruit blends and both work. Allulose gives a slightly softer set since it doesn't crystallize the way erythritol does, but the flavor is cleaner. Just adjust the amount based on your sweetener's equivalence to sugar. I stick with Swerve confectioners because the texture is the most consistent for me.
What other keto desserts pair well with these for a holiday spread?
I usually set these next to something chocolatey for contrast. My keto chocolate mousse takes about 10 minutes and gives you a completely different flavor profile. For fruit lovers, my keto strawberry pie is bright and fresh next to the warm pumpkin flavor.
Two years keto, pumpkin pie went on the closed list. I didn’t plan it, just stopped expecting it. A muffin tin and 5 net carbs changed that. Custard without the crust is the better version, honestly. Got four batches in since.
Made these last weekend and they were so good, but mine came out way softer in the center than I expected. Left them in the fridge overnight like you say and the outside set up fine, but the middle was almost like thick pudding. Not the firm custard I was going for. Pretty sure I measured everything right (eggs, heavy cream, the whole can of pumpkin, Swerve), so it has to be underbaking, right? They looked done on top when I pulled them out, but maybe my oven runs cool. I’ve also got a glass muffin pan instead of metal and I’ve heard glass heats differently, which maybe explains it. Either way I need to figure it out because the flavor was perfect and I’ve already made my husband eat three of them.
Glass, almost definitely. The edges set before the center does, so the top looks done when the middle is still liquid. Add 5 minutes next time and do the toothpick test – it should come out clean. Or switch to silicone, which is what I use now.
Good catch, Casey. Brand matters here. I use one that runs on the less-sweet side, which is probably why I landed on 1/2 cup. Better to start at a quarter cup and taste the batter. It should taste sweeter than you actually want, since the eggs and cream dial it back in the oven.
Stirred in half a teaspoon of orange zest because I was out of my usual spices, and it gave the custard this bright, lifted quality that cut through the richness. Wouldn’t have tried it otherwise but it’s the version I’m keeping.
Orange zest cutting through that custard makes more sense than spice does, actually. It goes after the fat in the cream in a way cinnamon can’t. Half a teaspoon is about right.
Didn’t think crustless pumpkin pie was worth making. Wrong. The overnight chill turns it into real custard. Four stars because I’m still adjusting to Swerve, but the texture got me.
I stirred in cinnamon before baking. Couldn’t help it, even though the recipe skips the spice entirely. Overnight in the fridge and it came out so smooth I kept checking if I’d actually messed something up. Flavor was warmer than expected, almost diner-pie territory. Now I need to make a plain batch to get what Annie means by ‘pure pumpkin custard.’
Tried subbing half the heavy cream with full-fat coconut cream for a dairy-sensitive friend, and the custard set up firmer, silkier somehow. Ended up preferring that version. Four batches in and I’m still doing the coconut swap.nnIf you’re using a metal muffin tin, run a thin offset spatula around each cup right out of the oven while still warm, then refrigerate overnight. They pop out clean the next morning, no tearing. The pumpkin flavor holds its own without any added spice, which I appreciate.
Coconut cream has way more fat than coconut milk, no surprise it sets up tighter. Four batches is a real answer. The offset spatula on warm metal works – I mostly just use silicone pans now.
Room temp eggs matter here, pulled mine out 30 min before and the custard was smoother across all 12 cavities. And the overnight rest is real. 4 hours and they won’t release as cleanly.
My grandmother used to make pumpkin custard in ramekins, no crust, just the filling. I always thought it was her thing, some old-fashioned shortcut. These taste exactly like that. The overnight set in the fridge, that silky dense texture with nothing to get in the way of the pumpkin. Two years keto and I stopped expecting recipes to actually hit like that. This one did.
One note for anyone who’s had these stick: I switched from a metal pan to a silicone muffin tray and they slide right out clean, no tearing. I also stirred in half a teaspoon of pumpkin pie spice and the spice level was much closer to an actual pumpkin pie. Refrigerated overnight and the texture completely changed, way more set and custard-like than straight from the oven.
Half a tsp is low enough it won’t change the macros much. And yeah, overnight is a completely different pie, I always tell people to make these the day before if they can.
If you use a silicone muffin pan, skip the cooking spray entirely. Mine released clean after 10 minutes in the fridge without any of the sticking I get with metal pans. Worth the swap if you have one.
Not sure how this is “pumpkin pie” without any spices in it. I’ve looked through the recipe, the printable but don’t see any spices in this recipe at all.
