Pinned Tweet
I wanted to make @BitRobotNetwork's open robotics lab brand mission a real experience that anyone can participate in
And today, that's possible with app.bitrobot.ai
Users can contribute directly to robotics R&D missions, earn rewards, and see their Access ID level up. Proud to develop this with the team. Check it out!
dabblerer retweeted
Thanks @KhalaResearch for this deep dive into robot learning and how industry leaders are tackling the data bottleneck.
Great to see our subnet partner @axisrobotics featured alongside BitRobot’s ecosystem crowdsourcing data across teleoperation, simulation, and ego video!
Robotics is still waiting for its "ChatGPT" moment
Language models had the internet to learn from, while robots need diverse physical demonstrations that are expensive to collect
Bessemer estimated just 300,000 hours of robot manipulation data globally in April, while others believe robotics could require 100M+ training hours
Bessemer projects $3B+ in robot-data spending over two years
Open networks are using crypto incentives to coordinate contributors across simulation, teleoperation and human video collection
Axis reports 7M+ trajectories from 220K+ contributors, with its data lifting task success from 83.9% to 88.8% on the LIBERO-Plus simulation benchmark
Our latest report explores the data opportunity through Axis, BitRobot, Eastworlds & THOR
Full report in the next post below
dabblerer retweeted
Robotics is still waiting for its "ChatGPT" moment
Language models had the internet to learn from, while robots need diverse physical demonstrations that are expensive to collect
Bessemer estimated just 300,000 hours of robot manipulation data globally in April, while others believe robotics could require 100M+ training hours
Bessemer projects $3B+ in robot-data spending over two years
Open networks are using crypto incentives to coordinate contributors across simulation, teleoperation and human video collection
Axis reports 7M+ trajectories from 220K+ contributors, with its data lifting task success from 83.9% to 88.8% on the LIBERO-Plus simulation benchmark
Our latest report explores the data opportunity through Axis, BitRobot, Eastworlds & THOR
Full report in the next post below
I'm a huge fan @atlasberry008 content, he is one of the few creators who has a deep pulse on the AI and tech industry and can break through X to reach a wider audience
Loved seeing this feature on the @BitRobotNetwork and @GroundedSI launch appear on my personal IG feed
BitRobot will pay you to wear a $500 hat at work.
Six cameras, pointed at your hands.
Robots have almost no training data, but people have all of it.
So the hats are going into bakeries, hotels and factories.
I would rather own what the humanoids learn from than fund the humanoids.
dabblerer retweeted
Thank you @atlasberry008 for this in-depth intro to the RoboCap platform with data enrichment from @GroundedSI
We're cooking new earning opportunities for the RoboCap subnet in the coming weeks. Stay tuned 👀
BitRobot will pay you to wear a $500 hat at work.
Six cameras, pointed at your hands.
Robots have almost no training data, but people have all of it.
So the hats are going into bakeries, hotels and factories.
I would rather own what the humanoids learn from than fund the humanoids.
dabblerer retweeted
The first Humanoid IKEA Assembly Challenge, co-organized with @UnitreeRobotics, @LightwheelAI and @singaporetech, took place at @ieeeiros.
9 global teams tested their loco-manipulation policies to assemble an IKEA UTTER children’s table. Full results ↓
dabblerer retweeted
The first ever Origami Challenge with @SharpaRobotics at @ieeeiros IROS 2026 concluded.
We asked teams to fold a paper airplane to test policies in dexterous manipulation.
14 global teams made it to the final round. Here are the results ↓
dabblerer retweeted
Better egocentric SLAM = better robot training data
GroundedSLAM leads every open system on LaMAria and offers real-scale tracking / mapping, sub-cm depth at manipulation range, and drift correction on long paths.
Congrats @GroundedSI on the release! Avail to RoboCap users only.
dabblerer retweeted
Day 2 @ieeeiros recap in the competition area
Crowds gathered to watch participating teams attempt new benchmarks in dexterous manipulation, long-horizon loco-manipulation, and auto-navigation.
Results from the 1st BitRobot Grand Challenges will be shared soon 👀
dabblerer retweeted
Oragami folding is quite an interesting benchmark test for robotic dexterous manipulation because it really challenges both the limits of hardware and AI model capabilities. Here they are using robot hand from @SharpaRobotics. I also learned from @micoolcho that there’s people in Japan that can fold a tiny crane as small as your finger tips all based on tactile and joint position with no vision at all. If robots can do that one day, then we know dexterous manipulation has been solved. Very impressive to see how much these teams can do in such short amount of time. Definitely not easy. Thanks @dabblerer_ for the tour! #IROS2026
dabblerer retweeted
Axis × BitRobot Alliance — Breaking Down the Latest Update
Following the latest update from the team, I want to break down what it actually means for contributors and what you should check on your account.
1. @axisrobotics has a fixed Bolts allocation
According to the latest update, Axis is allocated:
12,000 Bolts/day
360,000 Bolts/epoch
The important point here is that 360,000 Bolts is the allocation for the entire Axis subnet, not a fixed amount that is equally divided between every contributor.
Your individual reward depends on your share of the valid contribution made by all active contributors during that period.
A simplified example:
If your valid contribution represents 1% of the total valid Axis contribution, your share would be approximately 1% of the 360,000 Bolts allocation = 3,600 Bolts.
This is only a simplified example. The actual distribution can also depend on contribution quality, accepted points and applicable multipliers.
2. What does Pending Subnet Points mean?
This is probably the most important part of the latest update.
The reward flow can be understood as:
Valid contribution → Subnet Points → Bolts
If your dashboard shows Pending Subnet Points, it means your contribution has been recorded at the subnet level, but those points have not yet been converted into final Bolts.
So seeing Pending Subnet Points does not mean your Bolts are lost.
The points are waiting for the next distribution/conversion process.
Once processed, they should be reflected as Bolts on your profile.
3. What about the missing contribution issue?
The latest update also addresses the previous data upload issue.
Some contributor data was not uploaded completely, and the team has been working on this issue.
However, there is an important distinction:
The system issue being fixed does not necessarily mean every historical missing trajectory has already been restored.
The latest information does not confirm how many old trajectories were affected, how many have been restored, or whether every missing point has already been backfilled.
So if you previously completed contributions but your Subnet Points are still missing, you should check your individual account status rather than assuming everything has already been restored.
4. How can you understand your current status?
There are basically three situations:
🟡 Pending Subnet Points
Your contribution has been recognized at the subnet level.
→ Wait for the next distribution/conversion.
🟢 Bolts already appear on your profile
Your contribution has gone through the conversion process and the reward has been recorded.
🔴 No Subnet Points + No Pending Points + No Bolts
This needs further investigation.
Check whether the trajectory was uploaded correctly, whether it passed verification, whether it was rejected/duplicated, and whether your account and wallet are correctly connected.
5. Why can two contributors with similar activity receive different Bolts?
Because the system is based on contribution share, not simply the number of trajectories completed.
For example, if the total valid contribution increases because more people join or existing contributors produce more valid data, your percentage of the overall pool can change.
That's why:
More trajectories ≠ automatically more Bolts in a fixed ratio.
What matters is your contribution relative to the total valid contribution during the distribution period.
There can also be differences caused by contribution quality, accepted points and potential subnet multipliers.
6. Axis vs TeleArms
The latest information also gives us a useful comparison.
Both Axis and TeleArms have a stated base allocation of:
360,000 Bolts per 30-day epoch.
But this does not mean the same contribution will necessarily generate the same number of Bolts on both.
The result depends on:
• Active contributor count
• Total valid contribution volume
• Contribution quality
• Your share of the total
• Possible subnet multipliers
So the allocation is the same at the base level, while the individual distribution can still be different.
7. Don't forget the wallet connection
There is another important part to check if your contribution appears to be missing.
Axis uses a Base/EVM wallet experience, while BitRobot Connect uses the associated Solana wallet for contribution mapping.
Before creating a new account or wallet, check your existing setup:
→ Axis account
→ Axis → Subnets → Axis Robotics → Subnet Dashboard
→ Pending Subnet Points
→ BitRobot profile / Access ID
→ BitRobot Connect and wallet association
Creating multiple accounts or using the wrong wallet mapping can make it harder to trace your contributions.
8. What should you do now?
If you have Pending Subnet Points:
Keep a screenshot and wait for the next distribution.
If you completed a trajectory but have no Subnet Points:
Keep your trajectory/task ID, timestamp and relevant evidence, then check the verification status or submit a complete ticket.
If you have no task, no points and no Bolts:
Check your account, wallet and BitRobot Connect mapping first.
The key is to identify where your contribution currently sits in the process before assuming that a reward is missing.
One small suggestion for future epochs
It would also be useful if Axis could eventually provide an Alliance leaderboard for each epoch.
Not necessarily revealing everyone's exact numbers, but showing things like:
Your contribution → Your rank/percentile → Total valid contribution → Epoch progress
This would give contributors a much clearer view of where they stand relative to the overall contribution pool.
Since the final reward depends on contribution share, having this visibility would make the system much easier for users to understand.
For now, the main thing is simple:
Pending Points = recorded, waiting for conversion.
Bolts = processed reward.
No Points = something needs to be checked.
And the 360K Bolts/epoch is the total subnet allocation, not a fixed reward per user.
CC: @BitRobotNetwork @dabblerer_ @iamvella_
BitRobot × Axis Alliance — let’s talk about Bolts. 🦾
I’ve seen a lot of questions around Alliance Tasks lately, especially:
“Why did I complete tasks but not receive the Bolts I expected?”
The important thing is that Axis Points and BitRobot Bolts are not the same thing.
Think of it as:
Contribution → Subnet Points → relative share → Bolts
Alliance Tasks are built between Axis and BitRobot, so a valid contribution can create value across both systems.
But there is no fixed conversion like 1 Axis Point = X Bolts.
The important part is your relative contribution within the subnet.
A simplified example:
Subnet allocation = 100,000 Bolts
Total Subnet Points = 1,000,000
Your Subnet Points = 20,000
Your share:
20,000 / 1,000,000 = 2%
So your theoretical share would be around:
2% × 100,000 = 2,000 Bolts
This is only an example to explain the mechanism, not an actual allocation.
And this is why doing more tasks doesn't automatically mean more Bolts.
Imagine you increase your points from 10k → 15k.
Sounds good.
But if the whole subnet grows from 100k → 300k points, your relative share actually falls from 10% → 5%.
So what matters is not only how much you contribute, but also how much the rest of the subnet contributes during the same period.
Another important point: quality matters.
Rejected data, duplicates, low-quality trajectories or contributions that don't pass the required checks may not count toward the final calculation.
So personally, I wouldn't approach Alliance Tasks as:
“How many tasks can I complete?”
I'd rather think:
“How much valid, quality contribution can I add?”
You can also explore other BitRobot contribution gateways, such as Telearms, and compare how each subnet works.
Different subnets can have different allocations, contributor numbers and competition levels, so the relative contribution can be different.
But I wouldn't say one gateway will always give you more Bolts than another. The numbers can change between distribution periods.
The choice is yours.
Now, about the current Bolts issue.
BitRobot has just updated the community about the Axis × BitRobot Bolts App connection.
The team is actively troubleshooting with Axis and believes some data may be missing during the conversion.
Most importantly:
Your work and earnings are not lost.
So if you're currently seeing missing Bolts, incorrect numbers, or your contribution isn't reflected properly, please hang tight for the next update from the team.
Honestly, I'm also waiting for the latest update before suggesting what the next step should be.
I don't want to speculate while the team is still checking the data.
If you're having issues with Bolts, Subnet Points or Alliance Tasks, feel free to ask below.
I'll help clarify what I can, and for anything still under investigation, we'll wait for the official update from the team.
🦾 Keep contributing. Keep the quality high. And don't panic over missing Bolts while the data is being checked.
CC: @BitRobotNetwork VS @axisrobotics
Not only is this robot from @SharpaRobotics super cute, it’s also pushing the benchmark for robotic dexterity and fine manipulation by successfully folding a paper airplane.
Tmrw is the final day of @BitRobotNetwork competitions at @ieeeiros. Swing by to see the final results!
dabblerer retweeted
1/ Nobody has benchmarked a humanoid on autonomous IKEA furniture assembly before.
Today at @ieeeiros 2026, the first attempts begin.
It’s a test of autonomous loco-manipulation: long-horizon sequencing, precise coordination, contact, force, recovery, and movement ↓
Day 1 at IROS, impressive demo by @SharpaRobotics, where the robot beat the competing human most of the time catching sticks
The tech behind it: stereo cameras see falling stick → policy estimates 3D position + interception point → the end effector catches it and rapidly returns to its starting position
dabblerer retweeted
Starting next week, the Earth Rover Grand Challenges will be at @ieeeiros IROS 2026
11 teams will test their navigation policies in 4 tracks ranging from city to indoor / off-terrain to marathon (50 miles of sidewalks, overpasses and bike paths)
Let the competitions begin 🦾
Nobody has driven a mini rover from UC Berkeley to Stanford on its own.
Humans have done it by remote control, navigating 50 miles in 29 hours and 38 minutes.
That's the benchmark to beat at the Earth Rover Challenge co-organized by @BitRobotNetwork at @ieeeiros IROS 2026 ↓
dabblerer retweeted
Packed room at our office launch last Friday!
Builders from Physical Intelligence, Google DeepMind, Tesla, Anthropic, Dyna + beyond explored the full human data stack we’re building, from capture hardware to enrichment to training-ready outputs.
GSI was built to plug easily into the training flows teams already use.
dabblerer retweeted
Excited to co-organize the Origami Grand Challenge with @SharpaRobotics at @ieeeras IROS next week.
Sharpa builds advanced dexterous robotic hands and embodied AI models.
Over a dozen teams will put their policies to the test by autonomously folding a paper airplane.
Sharpa will be co-hosting two competitions at IROS 2026.
Powered by Sharpa's robot platforms and software, the 2nd ROCO and Origami challenges (cohosted by @BitRobotNetwork) brought together 48 teams from leading institutions around the world—including Carnegie Mellon University, UC Berkeley, KAIST and NUS.
Across both competitions, teams will put robotic manipulation and dexterity to the test on Sharpa hardware through challenging real-world tasks.
Follow the challenges for more:
ROCO: @NtuLab38456
Origami: robotic-origami-challenge.gi…
dabblerer retweeted
Next week, we’re heading to @ieee_ras_icra IROS 2026 in Pittsburgh with 50+ research teams pushing the frontier in 3 Grand Challenges:
> Autonomous navigation w/ Earth Rover
> Dexterity thru origami folding
> Loco-manipulation thru Ikea assembly
Excited to organize this w/ partners like @SharpaRobotics @LightwheelAI @UnitreeRobotics and more.
