SN3 Surpasses Market Gains as Bittensor's Decentralized AI Training Gains Traction

Templar (SN3), a Bittensor subnet, has completed the largest decentralized pre-training of a large language model (LLM) in history—Covenant-72B—with 72 billion parameters.
- The achievement demonstrates Bittensor’s capability to build competitive AI models in a decentralized environment, potentially shifting the valuation of TAO from narrative-driven to product-driven.
- NVIDIA CEO Jensen Huang likened Bittensor’s model to Folding@Home, highlighting the legitimacy of decentralized AI training and drawing attention from industry leaders and institutional investors.

Templar, a Bittensor subnet, has marked a pivotal development in the decentralized AI training landscape by completing the largest decentralized LLM pre-training run in history with Covenant-72B according to reports. The 72 billion parameter model was trained across a network of 70+ nodes using only standard internet infrastructure, showcasing the potential of decentralized AI infrastructure to rival centralized alternatives.

The success of Covenant-72B has led to renewed interest in Bittensor’s ecosystem. Institutional investors such as Grayscale and DCG have increased their exposure to TAO, the native token of Bittensor, reflecting growing confidence in the project’s technological and financial potential.

The market reactions have been strong. TAO surged 42.6% during the week of March 16, 2026, with SN3 tokens rising 75% in seven days, aligning with broader market trends that saw top 30 digital assets gain 8.8% on average. This surge indicates a shift in investor sentiment toward AI-driven blockchain projects.

What is the significance of Bittensor’s SN3 subnet?

Templar’s transition from data acquisition to model training represents a milestone in decentralized AI infrastructure. By utilizing a distributed network of nodes to pre-train Covenant-72B, the project demonstrated that decentralized systems can produce high-quality models without relying on expensive centralized resources.

The use of the SparseLoCo algorithm, which compresses communication between nodes by transmitting only 1%-3% of gradient components, reduces bandwidth needs and enhances the feasibility of decentralized model training. This technological advancement addresses a critical bottleneck in decentralized AI and positions Bittensor as a viable alternative to centralized training models.

How has the decentralized AI model influenced TAO and SN3 token valuation?

The completion of Covenant-72B has directly influenced the valuation of TAO and SN3 tokens. TAO’s 42.6% surge in the week of March 16 reflects a broader market reassessment of the token, driven by the subnet’s successful execution of its core training objectives.

The valuation shift is also attributed to the validation of Bittensor’s model by industry leaders like Jensen Huang, who compared it to Folding@Home. This endorsement has increased institutional confidence and highlighted the project’s potential to attract further investment and adoption.

What are the broader implications for the crypto and AI markets?

The success of Bittensor’s decentralized model has broader implications for both the crypto and AI markets. It demonstrates that decentralized networks can produce high-quality AI models, challenging the dominance of centralized players like NVIDIA and Meta. This shift could democratize access to AI training and reduce dependency on proprietary infrastructure.

The project’s institutional backing and market performance indicate a growing trend of integrating AI narratives into blockchain ecosystems. As more players explore the intersection of AI and blockchain, the landscape is likely to see increased innovation and competition, further legitimizing the crypto industry’s role in technological advancement.

Are there regulatory implications for this development?

The regulatory landscape for crypto assets is evolving, with the SEC recently declaring that most crypto assets, including staking and Bitcoin mining, are not securities. This guidance provides clarity for projects like Bittensor and reduces regulatory uncertainty for token holders and developers.

The SEC’s stance aligns with broader efforts to foster innovation while maintaining compliance with federal securities laws. However, the need for Congress to advance bipartisan legislation on market structure remains a key challenge, highlighting the importance of continued regulatory clarity for the sector’s growth.