const
@const_reborn
Building neurons, turning the web into a brain.
Missionary from the Church of RAO @opentensor
462 Following 31.1K Followers
This is the second subnet this week that has made physical world touch down (The other was @metanova_labs's connection to a automated drug lab)
This bridges the gap between the liquid intelligence layers on Bittensor, which exist purely in byte space, down into the hard atomic space where our systems can optimize and innovate harder products.
Very cool leap.
Show more
We've signed our first commercial robotics data pilot with Vulcan Shield Hungary, one of the largest advanced material manufacturing complex in Europe.
Starting at a production workstation in Békéscsaba, we'll combine real-world data and simulation to work toward a robot model trained for a specific factory task.
Show more
✅ BENCHMARKED: IOTA Decide 4B v0.4, an open-weight decision model from @MacrocosmosAI, running on my 64GB M5 Pro Mac mini
You hand it a state (a ticket, a policy, an agent trace) plus questions and one forward pass returns a probability for every answer. With their bundled loader on the Mac's GPU in full bf16, it loaded in 12.8 seconds and peaked at 10.7 GB.
Short ticket, about 130 tokens:
- yes/no: 88 ms
- 3-way choice: 105 ms
- 3-level score: 88 ms
That's about 11 decisions a second on a desktop Mac.
Their own example, a refund asked for 45 days after purchase on a 30-day policy, answered all 3 questions in 277 ms with the same answers on every repeat:
- eligible: 2.2% yes
- route: refunds team, 86%
- urgency: Low, 63%
Longer states take about 0.6 s per decision at 1k tokens and 2.6 to 3.0 s at 4k.
These numbers are the floor. On Mac its linear-attention layers still run on PyTorch's generic kernels, so an optimized Metal path should make it faster. Their 33 ms figure is on a dedicated GPU.
If you're building agents, about 90 ms a decision is fast enough to check a policy before an agent acts or route tickets on your own machine, with no per-call bill. I didn't benchmark accuracy, but it gave the same answers on their refund example every run.
Show more
The reason people reacted against centralized data centers is that they need to feel like they own a piece of the future
Decentralized AI and personal superintelligence is the way forward
A future where people own and control the extension to their cognition, both hardware and software, is needed
Show more
4* frontier model releases from Bittensor today.
ChronoLLM-16B pre-training is almost done and was built on Bittensor, by Subnet 38.
2 big updates:
1. ChronoLLM will release the largest Point-In-Time LLM out there, a 16B MoE.
2. Early vintages beat every point-in-time model at every cutoff through 2020. ChronoLLM should own the frontier for LLMs without lookahead.
ChronoLLM-16B is 8x bigger than its predecessor and scores 22% higher.
Our closest comparable is AQR's point-in-time model, built by a $100B asset manager. ChronoLLM-16B has 4x the parameters, 8x larger context window and scores 12.5% higher on the Open LLM benchmark (22% on the 2013 cutoff). For scale, 10% on this benchmark is the gap between a 7B and a 13B Llama: a full model-size tier. And it stays flat across every cutoff from 2013 to 2022. That’s what no lookahead looks like to us.
ChronoLLM-16B is also the first point-in-time model that can reason. Until now, every other point-in-time model scored near random on reasoning benchmarks like MMLU and GSM8K. ChronoLLM doesn't.
We’re here to stay.
Show more
Kimi K3.1 Pro is now live on Engy, 30% off! 🚀
Per 1M tokens:
• Input: $1.40
• Output: $7.00
• Cached input: $0.14
Text + image input. Pay as you go.
Try it →
3 pareto-frontier models released on Bittensor today.
Introducing IOTA Decide 4B, our first public model.
Pareto optimal. Built different.
Trained across globally distributed GPUs with our SDK.
Open weights, Apache-2.0.
🤗
A small team on Bittensor just beat the big labs.
HaloGuard 2.0 is the new SOTA for open streaming guards: ahead of Qwen3Guard-Stream at every size. At 0.8B.
Bittensor isn’t about following the frontier. It’s about defining it.
Show more
recursive auto research swarms feeding autonomous drug labs, 2026
accelerating and ultimately automating drug discovery depends on autonomous robotics labs like @Ginkgo has built.
their Boston facility is already the world's largest.
wish i could've joined!
Everyone waking up to the idea of decentralized inference networks now, but also compute clusters, and training systems.
It will be protocols that mediate those networks since only a protocol can aggregate disparate and liquid compute -- across jurisdictions -- without taking on localized political risk.
gn datacenters --> gm inference networks
@engyai @chutes_ai @thinkinstant
Show more
whether AI fascism or AI doom fanaticism, the answer is still in cyberspace.
1 PB stored on Hippius.
1,000 TB of data, held on a distributed network.
Thank you to everyone who trusted us with their data. This is just the beginning.
Show more
Eager to see the #USElection# midterm results?
We’ve used state-of-the-art AI models to forecast the outcome.
With Ephemeris, our forecasting inference product, we’ve projected the Senate results:
Via zero-shot forecasts, our model tells us the Democrats are leaning to 53 seats.
Leans D: Kansas +4.6, Ohio +3.9, Maine +3.8, Texas +2.8, Alaska +0.8
Leans R: Iowa +0.8, South Carolina +1.0
Coin flip: Louisiana
We'll score them tonight.
Show more
📸 Current Zeus leaderboard snapshot!
Four UIDs capture ~95% of emissions, none of them dominate every single variable.
🏆 UID 224 (#1#) wins temperature with solid wind performance to hold the lead.
UID 85 (#2#) absolutely dominates solar, but capped by its 20% scoring weight.
UID 229 (#3#) is crushing wind. Set to climb fast with wind's current 60% scoring weight.
🔗
Show more
the real* up and to the right charts
100 steps of fully decentralized RL on Reliquary.
Rollouts from untrusted miners on 12+ GPU models. That usually breaks RL: we fix the sampling noise publicly, so the trainer knows exactly what the model would have sampled.
Every eval sample verified on-subnet.
bittensor:native
Show more
Today Numinous is releasing its impact UI!
Built from our underlying causal graph, it shows for each equities their exposures along macro and geopolitics factors, tracked by prediction markets. As the event landscape changes, your fund can track the mechanisms propagating to equities.
Show more
Welcome to the era to decentralized post trained models that beat the frontier in their classes.
Introducing Reliquary-4B.
A 4B math & code model trained with reinforcement learning. Anyone could join the network and contribute rollouts.
Independent miners chose the prompts and generated the rollouts. The protocol verified them and trained the model.
Here’s the model and the research behind it.
Show more
Albedo Agent API is live for free.
Miners get 10x standard quota - resets every 24h.
Drop a key into Claude Code, Codex, Copilot, Cursor, ACP.
Sign in with GitHub, copy a key, paste it into your agent.
Use it on real work and LMK how it feels.
Show more
The Hippius mobile app is live on Android.
Your photos and files, backed up automatically and end-to-end encrypted on your phone before they ever upload.
The same account and storage you already use on desktop and web. Your whole cloud, now in your pocket.
iOS is coming next.
Show more
I love the push towards private AI, but a seperate source of freedom will be truly public AI, i.e. a world where the majority of AI trajectories are searchable, and thus minable for us to improve, understand and monitor intelligence -- not just the central labs.
Show more
