beincrypto.com 
Bittensor ($TAO) surged 19.19% in the last 24 hours, fueled by a wave of demand tied to its AI-powered subnet ecosystem.
The rally coincided with a viral social media moment from Templar, one of $TAO’s most active subnets, and growing interest in decentralized artificial intelligence tokens.
Subnet Token Demand Drives $TAO Price Higher
As of this writing, Bittensor’s powering token, $TAO, was trading for to trade at $284.75, up by 19.19% in the last 24 hours.
The catalyst behind $TAO’s move may be straightforward. Investors need to hold $TAO to exchange into subnet tokens, and a surge in Templar hype created a wave of buying pressure.
Tao Telegraph, a community account, noted that a single viral Templar post was enough to push $TAO demand higher, speculating that simultaneous hype across multiple subnets could amplify this effect significantly.
Let’s be real, $TAO is up because of this viral post by @tplr_ai
— $TAO Telegraph — Bittensor Media (@taotelegraph) March 14, 2026
To buy tao subnet tokens, investors NEED to own $TAO to exchange into it… there is no way around this.
Hype for one subnet token this week caused $TAO demand to rise, just imagine if there is this type of demand… https://t.co/d8x4zGlpuy
Three $TAO ecosystem subnets, SN3 Templar, SN4 Targon, and SN39 Basilica, ranked among the top eight daily gainers according to CoinGecko. $TAO itself broke above $280 during the move.
$TAO breaks above $280 as three of its ecosystem subnets rank among the top 8 daily gainers.
— CoinGecko (@coingecko) March 15, 2026
🟢 SN3 – @tplr_ai
🟢 SN4 – @TargonCompute
🟢 SN39 – @basilic_ai pic.twitter.com/awtKkpW6ZG
Templar’s Covenant-72B Marks AI Milestone
Behind the hype sits a tangible technical achievement. On March 10, Templar announced the completion of Covenant-72B, a 72-billion-parameter Large Language Model (LLM) pre-trained entirely on Bittensor’s Subnet 3.
The model was trained on roughly 1.1 trillion tokens using commodity internet connections. No centralized cluster or whitelist was involved. Anyone with GPUs could participate freely.
Templar used a technique called SparseLoCo to overcome bandwidth limitations. Each participant ran local optimizer steps before compressing and sharing updates, making decentralized training feasible at 72B scale.
We just completed the largest decentralised LLM pre-training run in history: Covenant-72B. Permissionless, on Bittensor subnet 3.
— templar (@tplr_ai) March 10, 2026
72B parameters. ~1.1T tokens. Commodity internet. No centralized cluster. No whitelist. Anyone with GPUs could join or leave freely.
1/n pic.twitter.com/W0Ks563Cld
The team said Covenant-72B delivers performance competitive with centralized models like LLaMA-2-70B.
Van de Poppe Rebalances Around $TAO Strength
Trader Michael van de Poppe disclosed that he sold 10.42 $TAO at $288 for roughly $3,000. He rotated proceeds into SEI and EIGEN while keeping nearly 50% of his portfolio in $TAO and NEAR.
“The AI <> Crypto narrative is one of the core angles of the upcoming markets,” wrote Van de Poppe.
He flagged $TAO’s move as a 1.8 sigma event on the daily timeframe, calling it slightly overextended and ripe for a short-term correction that could offer a re-entry point.
Whether $TAO can sustain these levels may depend on continued subnet activity and broader AI sector momentum heading into next week.
The post Bittensor ($TAO) Surges 20% as Templar’s Viral Subnet Hype Fuels Buying Frenzy appeared first on BeInCrypto.
