MicroDuck is exactly the kind of product that can make an impulse purchase feel reasonable.
You see it walk.
Then it sits down.
Then it grabs something.
Then you see the rollers.
Then someone mentions that the software stack is open source.
And then you see the price:
$399 before taxes and shipping.
That is a surprisingly accessible price for a small biped robot.
But a preorder is still a preorder.
Before spending the money, I would separate three things:
- what MicroDuck can do today
- what you are actually getting for $399
- what you are buying because you believe the platform could become more capable later
Those are not the same thing.
1. Start With What Is Actually in the Box
Pollen's current standard MicroDuck package includes:
- the MicroDuck robot
- a battery
- a USB-C cable
- a game controller
That is enough to get started with the robot itself.
But some of the things you may have seen in demonstrations are sold separately.
Pollen currently lists three additional packs:
- Charger Pack — $39
- Dev Pack — $119
- Accessory Pack — $39
The Charger Pack includes a dual charger and two additional batteries.
The Dev Pack includes spare motors, motor cables, batteries, a dual charger, NFC tags, a screwdriver, screws, and other development-related items.
The Accessory Pack includes items such as the roller attachments, ball, laser pointer, NFC accessories, and additional NFC tags. Pollen lists these separately from the base robot.
That distinction matters.
If the thing that made you want MicroDuck was a roller-skating demonstration, do not assume the rollers are part of the base package.
Before ordering, check the current product page and make sure you understand exactly which version of the product and which accessories are in your cart.
2. The $399 Price Is Not Your Final Cost
The advertised introductory price is currently:
$399 before taxes and shipping.
That is the important number to remember.
Not:
"$399 all-in."
Depending on where you live, taxes and shipping can increase the final amount.
And then there are optional packs.
You could very quickly turn:
$399 robot
into:
$438 robot + charger
or:
$478 robot + accessories
or:
$597 robot + development and accessory packs.
That does not mean those packs are bad value.
It means you should not automatically buy all of them because the product page makes them look interesting.
The base robot is already intended to be playable out of the box.
I would start there unless you already know why you need the additional hardware.
3. Understand What You Are Ordering: A Preorder, Not a Shelf Product
This is probably the biggest practical difference between MicroDuck and buying an ordinary consumer gadget.
Pollen currently lists MicroDuck as a preorder.
Its official product information says the introductory price is $399 before taxes and shipping, with first deliveries targeted before Christmas 2026.
That is a target, not the same thing as having a robot sitting in a warehouse ready to ship tomorrow.
So I would not make a purchase decision based entirely on the date you saw in an old post, video, screenshot, or launch announcement.
Check the current ordering page when you actually place the order.
The delivery expectation is part of the purchase.
If you are buying because you want a small robot to play with immediately, a preorder may be the wrong format for you.
If you are comfortable waiting for the product and watching it develop, that is a different decision.
4. Decide Whether the Robot Is Interesting Without the Future
This is the test I would take most seriously.
Imagine that MicroDuck never becomes dramatically more capable than it is today.
Forget the future AI companion.
Forget some hypothetical home assistant.
Forget a huge community of new skills.
Forget the idea that every open-source robot eventually becomes much more powerful.
Now look at the robot as it exists.
Pollen currently lists seven trained behaviors in the box, including:
- walking
- sitting and standing
- kicking
- grasping
- roller skating
- getting back up after a fall
It can also be manually controlled with the included game controller.
That is already a real product.
The question is whether that product is enough for you.
If you look at those capabilities and think:
"That already sounds fun."
Then the preorder has a solid foundation.
If you think:
"It will only be worth it if they eventually add ten more features."
I would be much more cautious.
You should like the robot you are buying.
You should not need the future version of the robot to justify the current purchase.
5. Be Precise About What "Open Source" Means
"Open source" is one of the most attractive parts of MicroDuck.
It is also one of the easiest things to misunderstand.
Pollen currently says that the SDK, simulation environment, and full reinforcement-learning training stack are available openly. The software stack is published under Apache-2.0, and the seven shipped behaviors are published as retrainable policies.
That is meaningful.
It means MicroDuck is not simply a closed appliance where the manufacturer's software is the end of the road.
You can inspect the software.
You can work with the simulation.
You can retrain behaviors.
You can potentially create something new and deploy it to the robot.
But this does not mean:
"Everything about the robot is open source."
It also does not mean:
"Anyone can instantly teach it any task."
There is still a gap between having an open robotics stack and having the skills, time, and hardware needed to use that stack effectively.
So ask yourself what you actually mean when you say you want an open robot.
If you mean:
"I want to explore and modify the robot's software."
MicroDuck is genuinely interesting.
If you mean:
"I want to freely manufacture every part of the robot myself."
That is a different proposition.
6. Decide Which Kind of Owner You Are
I would put potential MicroDuck owners into two broad groups.
The player
You want a small physical robot that is fun to control.
You want to see it walk around.
You want to try its behaviors.
You want something physical rather than another application on a screen.
You may never touch the development stack.
That is completely fine.
The builder
You are interested in what happens underneath.
You want to work with simulation.
You want to train policies.
You want to experiment with reinforcement learning.
You want to modify behaviors.
You care about the fact that the software is open.
This is where the platform side of MicroDuck becomes especially interesting.
The important thing is that you do not need to pretend to be the second type of owner if you are actually the first.
You can buy MicroDuck simply because you enjoy robots.
7. Make Sure Your Space Matches the Product
MicroDuck is only about 25 cm tall and weighs less than 800 g.
That makes it small compared with many robots.
But "small" does not mean "space-independent."
It is still a physical machine.
It needs somewhere to move.
So think about your actual environment.
On your desk
A desk is a natural place to keep it.
But a desk also creates obvious constraints.
A walking robot needs enough surface area to move without immediately reaching the edge.
On the floor
You get more room.
But now there are other problems.
Furniture.
Cables.
People walking past.
Pets.
Uneven surfaces.
Things that can knock it over.
Around other people
This can actually be part of the fun.
Pollen's own launch material shows MicroDucks being used around people and in shared activities.
But that also means the robot is not just a stationary personal gadget.
It becomes an object moving through a physical environment.
That is worth considering before buying one.
8. Know What You Are Actually Buying Into
There are really two purchases here.
The first is:
the MicroDuck that exists today.
The second is:
the possibility of a more capable open robot platform later.
The first is tangible.
The second is uncertain.
That distinction matters because the second part can be very tempting.
Pollen's current software stack is built around a clear train-simulate-deploy-refine workflow.
You can train behavior in simulation, deploy it to the physical robot, refine it, and publish the resulting policy.
That is a serious proposition for a $399 consumer robot.
But the existence of the workflow does not guarantee that you personally will use it.
And it does not guarantee how large or active the future community will become.
Treat that part as upside.
Do not treat it as something you have already bought.
What I Would Check Before Clicking Preorder
At this point, I would ask myself eight very simple questions:
1. Do I like the robot as it exists today?
Not the roadmap.
Not the launch video.
The actual current robot.
2. Am I comfortable with a preorder?
I am not buying an immediately available mass-market gadget.
3. Is $399 comfortable before taxes and shipping?
Not "can I technically afford $399?"
I mean:
"Would I still feel good about this purchase if the robot took longer to arrive than I expected?"
4. Do I actually want a physical robot?
If what I really want is an AI conversation, there are cheaper and simpler ways to get that.
MicroDuck's entire reason for existing is partly that it is physical.
5. Will I use the built-in behaviors?
If walking, grasping, kicking, rolling, and physical interaction already sound fun, that is a strong signal.
6. Do I care about the open software side?
If yes, the platform becomes much more interesting.
If no, you are mostly buying a playful physical robot.
7. Do I have somewhere to use it?
A robot that spends 99% of its life sitting in its box is not a good purchase.
8. Am I buying the present or the future?
This is the most important one.
If you are buying because you like what exists now, good.
If you are buying because you need several future capabilities to arrive before the product feels worthwhile, be careful.
Who I Think Should Preorder
I think MicroDuck is particularly interesting for someone who says:
"I like small robots, I want something physical to play with, and I am intrigued by the idea that I could go deeper into the software later."
That is a very good fit.
It is also interesting for developers and robotics enthusiasts who specifically value an open simulation and reinforcement-learning workflow.
The $399 starting price makes the entry point unusual compared with many more expensive research-oriented robots.
But the price only looks compelling if the product category itself interests you.
Cheap for a robot is still expensive for a product you do not actually want.
Who I Would Tell to Wait
I would wait if your expectations are mainly:
"I want a fully autonomous AI assistant."
Or:
"I want a robot that understands my home."
Or:
"I want something that is immediately useful every day."
Or:
"I want a conversational companion with deep memory."
Those are different product expectations.
MicroDuck may eventually become part of a broader ecosystem around those ideas.
But that is not the reason I would justify a preorder today.
My Take
I think MicroDuck is one of the more interesting $399 robot propositions precisely because it sits between two categories.
It is playful enough to be approachable.
But it is also open enough to be interesting to people who want to build.
Pollen currently describes it as a 25 cm open-source biped, playable out of the box and designed to be retrained through reinforcement learning. The base package includes the robot, battery, USB-C cable, and controller; additional packs are optional. (Pollen Robotics)
That is a compelling combination.
But I would still make the decision using a very simple rule:
Would I be happy owning the MicroDuck that exists today, even if the future turns out to be less ambitious than I hope?
If the answer is yes, I think the preorder makes sense.
If the answer is no, I would wait.
Because you are not really buying a promise.
You are buying a small robot.
The promise is the bonus.
