Stillcore Capital

Hash Rate - Ep. 180: Stillcore 'State of Subnets' Report 🧙 Guest: @markjeffrey of @stillcorecap A guided walk-through of the latest Stillcore report on Bittensor $TAO Subnets.

Presenting: The Stillcore Capital - STATE OF SUBNETS July 2026 Report Download: docsend.com/view/562fmtiey…

Need. More. Tokens. Testing a secret new sub on @bittensor network running GLM 5.2 and it’s mind blowing 🤯 when your token cost drops to zero

I am pleased to announce that Stillcore Capital @stillcorecap has invested in Minos @theminos_ai (Bittensor $TAO Subnet 107)

Exciting to see Chamath and J-Cal discussing the power of Targon distributed confidential compute on @theallinpod “I do think this idea of distributed inference has a real place in the American ecosystem.” We appreciate the mention, and are excited to watch the continued growth of the permissionless compute space.

A more detailed answer to @travisk's objection on the latest @theallinpod that decentralized training of AI is not doable because of bandwidth: Traditional data-center training uses high-bandwidth interconnects (NVLink/Infiniband) and keeps everything co-located because communication is a huge bottleneck. IOTA (Bittensor Subnet 9) makes decentralized training feasible through several innovations: - Memory scaling: No single device needs the full model in VRAM. Capacity scales with the sum of participants’ memory, not limited by the weakest one. - Communication efficiency: Heavy activation compression + efficient all-reduce + concurrent compute/communication + stochastic routing for fault tolerance. The run achieved >30% MFU (peak sustained ~38%), roughly 65% of comparable co-located datacenter speed. - Fault tolerance & decentralization: Handles churn, variable bandwidth/latency, and unreliable peers via thresholds, redundancy in all-reduce, and randomized pathways. An orchestrator coordinates high-level triggers, but actual compute and much of the data flow is peer-to-peer. - Incentives: Bittensor’s blockchain layer rewards contributors fairly and granularly based on verifiable work, discouraging hoarding or free-riding. - Practical results: Orion-100B trained ~1.1 billion tokens over ~2 days on these 48 distributed GPUs at a much lower cost than centralized. This demonstrates that frontier-scale training over commodity internet is becoming technically and economically viable.

Excited to announce that Stillcore Capital @stillcorecap has invested in Aurelius @AureliusAligned (Bittensor Subnet 37)

Hash Rate - Ep. 176: Macrocosmos IOTA (9) + APEX (1) 🧙 Guests: @WSquires and @macrocrux of @IOTA_SN9 and @Apex_SN1 Download SN9's Home Mac Miner: iota.macrocosmos.ai 00:00 Orion 100B Model 03:27 Comparing Training Approaches 10:15 Liquid Training 12:17 IOTA: One-Click Mining 25:58 The Journey of IOTA: A Year in Review 33:04 Training Models: The Lifecycle and Challenges 36:06 Balancing Training and Inference in AI 37:56 AI: Scaling and Diminishing Returns 43:00 Aurelius: New Partnership 58:55 Upcoming Events and Community Engagement

🔥

bittensor:native is the printing press of intelligence. 21 million tokens mined with useful ai work. the world will tap into Bittensor’s decentralized compute & intelligence like a faucet intelligence will be commoditized the way information was after gutenberg this is one of the greatest asymmetric opportunities of all time

reed’s law on TAO is closer to a triple exponential. subnets are group-forming coalitions (2^n). they compose with each other. and the agents inside them improve recursively, earning emissions to fund their own compute. 128 subnets today, thousands coming. this is the decentralization of intelligence itself, biggest story in crypto since bitcoin, and the market hasn’t priced any of it. the next iteration of the internet is being built on bittensor. 100% compatible with your everything code. we should do a pod sometime

Great read

WOAH @Jason discussing @ridges_ai and Bittensor on @theallinpod 👀 there's clear opportunity post-templar to be the king subnet on $tao. chat, i need your help... is @ridges_ai the next $100M subnet runner and taking the subnet crown?

Jason’s on Bittensor (TAO): Bittensor is a marketplace for AI services. People compete to provide a service… Things like compute and models. The incentive structure is brutally deflationary: better quality + lower price wins. If buyers don’t pay enough, miners leave. If the work is priced right, the best talent shows up. @jason #Bittensor #TAO #Crypto #AI #DePIN

Nice mention & discussion of Ridges @ridges_ai sn62 / Bittensor on the ALL-IN pod that just dropped: youtu.be/DVBJQQCjgXU?si…

Exploits are what teach a system its weak spots. The quicker you find them the faster you learn. The outcome of this eventful evening is that Bittensor will invent lock-based subnet ownership -- specifically: ownership of a subnet determined by a team's long term economic commitment to the project. This will mean: 1) investors see long in advance if an owner has unlocked their tokens, 2) be able to reprice the subnet before the owner and 3) liquidly direct their own conviction to another team, or agent, to manage the system. Thank you @DistStateAndMe for helping further Bittensor's decentralization and develop a solution to one of cryptos oldest problems: founders who rug their token holders. Looking forward to training some 1T param models with the miners who are experts in this unique field. "What is dead can never die"

I am pleased to announce that Stillcore Capital @stillcorecap has invested in Synth @SynthdataCo (Bittensor Subnet 50).