HomeInsidersWhalesCohortsAllocationsLiquidityThe TableNewsEmissionsX-RaySmart MoneyOutflowsData APIPricing
● Live insider tape…
Subnets / SN124

Swarm SN124

Swarm (subnet 124) is a Bittensor subnet held by 2,082 unique coldkeys. Its liquidity pool holds 13.0k τ, with the alpha token trading at 0.0112 τ. Roughly 77% of tracked holders are in profit — the holder base is broadly in the green. Seller exhaustion is elevated: 81% of recent sellers have emptied their wallets, thinning the supply overhang. Full insider flow, whale tracking and the complete proprietary factor set for Swarm are available to SubnetStats subscribers.
SCREENER
ALPHA PRICE
0.0112 τ
IN TAO
HOLDERS
2,082
UNIQUE COLDKEYS
POOL DEPTH
13.0k τ
TAO IN POOL
7D MOMENTUM
-3.6%
ALPHA vs TAO
01

30-day alpha price

as of 2026-09-15
Project activity
@swarmsubnet
6h ago
Our drone's camera is 256 pixels wide. At that size a door renders as one flat shape, its edges lost in the wall. We are changing how surfaces get drawn, so a door stays a door. https://t.co/DswXwUgCW0
1d ago
Langostino, our open-source autonomous drone, is closing in on 500 GitHub stars 🚀 At Swarm, we believe intelligence shouldn’t stay locked inside a lab. Link in the first comment. Build your own autonomous drone like it’s an IKEA wardrobe. https://t.co/nooff20QMw
2d ago
Every seed gets its own sun ☀️. One mountain, one day. The hour is drawn from the seed on a real solar arc, so one flight starts in flat dawn light and the next one at hard noon. You cannot memorise the light. https://t.co/Xp014qjFCi
3d ago
We are rebuilding what our drone sees. Our simulator paints the sky white. Every frame, every map. Any model that learns from that picture is learning a world that does not exist. The sky comes first 🌄. https://t.co/Y8nDOe0obn
4d ago
SWARM just landed its second commercial project. Mevi Dron picked us to build an avalanche-prevention drone, starting with Andorra's biggest ski resort. The mission: make controlled avalanche release safer and more efficient. The SWARM lab is at work. More below 👇 https://t.c
6d ago
Come meet the team at @ExploitSummit!
7d ago
We are building an open-source drone interceptor and testing it in our real lab. Our simulated lab now rearranges itself. Same room, same 41 objects, a new layout every run. A memorised map no longer matches the room. Infinite labs. One real one. The interceptor is coming. http
8d ago
Swarm submissions are now private. Your model stays hidden while it competes. Nobody can copy it. Champions still go public: full source and weights, the moment they take the crown. https://t.co/BqabFDbL4m
11d ago
An indoor drone hunting another drone. No GPS. No compass. One camera, a room it has to search, and a target that runs. Five missions, five catches — every one a real evaluation episode. The seed is on screen, so anyone can fly it again. https://t.co/Q6gv8jgo2D
14d ago
Swarm has begun developing the simulations behind an autonomous surveillance drone for solar parks, under our commercial agreement with German Solar Power. A key step toward truly client-tailored, applied intelligence. https://t.co/OAY0yUpaJU
15d ago
Swarm Miner Survey #1 is live. You train the models. You compete. You know where mining works—and where it doesn’t. We want your honest view: what should stay, what should change and what we should improve next. Help shape Swarm. Take the survey: https://t.co/QYUCpwUeEk
16d ago
Reminder: Apply now if you want to build the future of autonomy
17d ago
Hand Bittensor miners a problem nobody has solved yet, and this is what one day looks like. 35 models in the first 24 hours, one every 41 minutes, overnight included. The top score was beaten 11 times: 0.60 → 0.9085. It stands at 0.9763 now. Indoor drone pursuit. No GPS, no com
19d ago
We’re hiring an Autonomy ML Engineer at SWARM. You’ll work on reinforcement learning for autonomous drones: navigation, obstacle avoidance, mission execution, sim-to-real, and increasingly capable flying agents. Python + PyTorch + RL + robotics. Based in Andorra. Apply: https
20d ago
3 months later: ✅ Lab built ✅ Exceptional CTO onboarded ✅ Benchmark: 5× bigger ✅ Autonomous flying agent: live contracts Promise. Execute. Deliver.
21d ago
An interceptor wins by making contact. Success shouldn’t mean a broken drone for our lab So we 3D-printed a protective frame for testing it Printed ✅ Installed ✅ Flight-tested ✅ Build. Test. Iterate. Repeat That’s the speed of a physical lab https://t.co/yew5FYVc4b
22d ago
Today we launch our first open competition to train a drone interceptor with real hardware waiting for the winner Original Interceptor hit 100%🐝. Warm-up over. Built for sim-to-real transfer. We recreated our lab and drones in simulation. We’re going for the home run https://t
24d ago
Every epoch, Swarm generates 1,100 environments that have never existed before. They are built the moment the epoch opens and destroyed when it closes. There is nothing to memorise. https://t.co/nlmqdrIiam
26d ago
https://t.co/7UHTiZ6KiU
28d ago
SWARM has landed its first commercial project German Solar Power selected us to build autonomous drone surveillance for solar parks in Greece. The mission: detect intruders, fires, hotspots and equipment failures—then dock, recharge and patrol again Autonomy, put to work. http
2026-08-12 ago
We built Arbuzz to support the best network of miners in Bittensor. Since May it has answered 1,288 of their questions, 97.3% of them without a human ever stepping in. At every hour of the day. Our miners are everywhere.
2026-08-11 ago
This was only one year ago. One video says more than we could ever summarize in a post https://t.co/3y9Xn6Dyj1
2026-08-11 ago
Happy to announce that our CTO will also be joining us at @ExploitSummit! Come meet her and get a deeper understanding of our team, our skills, the challenges we face, and what we’re building. bittensor:native
2026-08-10 ago
789 models entered SWARM. Five hold a crown. As of today, all five champions are public. Take the strongest autonomous drone models miners have built. Run one. Fork one. Beat one. https://t.co/zofT16OSou
2026-08-07 ago
This drone exists twice One flies in our simulator. The other is a real Tello waiting in our lab Same dimensions. Same weight. Same rotor layout We built its digital twin so models can train in simulation, then graduate to real flight https://t.co/8euPhnpbH8
Posts by the project. Not a signal.
02

Proprietary signals

a sample — full set on the screener
SELLERS DRAINED
81%
28D SELLERS EMPTIED OUT
CAPITULATION
-0.21
BOUNCE COMPOSITE (z)
HEAT 7D
0.35×
VOLUME ÷ POOL DEPTH
HOLDERS IN PROFIT
77%
TRACKED WALLETS · ENTRY BELOW PRICE
See who's buying and selling Swarm — every trade classified miner / validator / owner / outsider, plus holder cost basis and the full 23-factor screener.
OPEN THE INSIDER TAPE →
← SN123All subnetsSN125 →
Figures computed from Bittensor chain records. Not investment advice.
Bittensor subnet intelligence · Powered by TAOstats API · Methodology v1.0 · Not investment advice© SubnetStats · @SubnetStats · Feedback