After Institutional Support and Price Surge, Revisiting the True Value of Bittensor's 128 Subnets

marsbitPublished on 2026-03-17Last updated on 2026-03-17

Abstract

After removing institutional support and price increases, this article re-evaluates the real value of Bittensor's 128 subnets. Bittensor operates as a decentralized AI ecosystem where each subnet functions like an independent startup with its own token (Alpha), revenue model, and team. There are two primary ways to earn: TAO emissions (protocol subsidies based on staking inflows) and Alpha token PnL (capital gains from subnet performance). Since the Taoflow update in November 2025, subnets with negative net staking flow receive zero emissions, creating a competitive environment. Approximately 3,600 TAO (around $960k daily) is distributed, with the top 10 subnets controlling 56% of emissions. Key case studies include Chutes (SN64), which demonstrates product-market fit with 400k users and 9.1 trillion tokens processed at 85% lower cost than AWS, and Templar (SN3), which offers asymmetric upside by training frontier LLMs in a fully decentralized manner. The investment framework positions TAO as an index fund for the entire network, while Alpha staking represents concentrated bets on specific subnets. The ecosystem is attracting institutional interest, with significant holdings from DCG and Polychain Capital. The conclusion emphasizes evaluating subnets based on product utility, staking flow, team execution, organic demand, and liquidity conditions.

Author: Kaff

Compiled by: Yuliya, PANews

TL;DR

  • Bittensor consists of 128 independent subnets, each operating like a startup with its own token (Alpha), revenue model, and team.

  • There are two ways to make money: TAO emissions (protocol subsidies from staking inflows) and Alpha token PnL (capital gains from subnet performance).

  • Since Taoflow in November 2025, subnets with negative net staking inflows receive zero emissions—forcing them to either succeed or die.

  • Approximately 3,600 TAO (~$960k) is distributed daily across all subnets, with the top ten controlling ~56% of the share.

  • Chutes (SN64) is the clearest example of product-market fit: 400k+ users, 9.1T tokens processed, 85% cheaper than AWS.

  • Templar (SN3) is the most asymmetric bet: fully decentralized training of frontier LLMs with a ~$60M market cap vs. OpenAI’s $800B.

  • TAO = broad index fund to the entire network, while Alpha staking = concentrated investment in specific startups—with 100%+ APY potential but real risk.

  • Alpha tokens carry no formal yield promise; their value depends entirely on market dynamics and team execution.

1. Subnet Structure: Who Does What?

When people think of Bittensor, the most common mental model is: this is a decentralized AI project. While true, it’s incomplete.

In reality, Bittensor is 128 independent AI startups competing in a brutal economic system, each with its own token, revenue model, and struggle for survival. By March 2026, the total market cap of all subnet tokens was ~$1.12B, equivalent to 27% of TAO’s own market cap. Grayscale calls it “Y Combinator for decentralized AI.” Except instead of a committee deciding who gets funded, the market decides.

Understanding this mechanism allows you to evaluate which subnets are creating real value and which are dying.

Each subnet is an incentive-based competitive market producing specific digital goods—AI inference, GPU compute, model training, financial data analysis, and more.

Each subnet is driven by three roles: subnet owner, validators, and miners:

Alpha Tokens: Equity in the Subnet

When you stake TAO to a subnet, your TAO enters an on-chain AMM pool (similar to Uniswap V2 mechanics). In return, you receive Alpha tokens. The price formula is:

Alpha Price = TAO in Pool ÷ Alpha in Pool

Alpha tokens have a hard cap of 21M (echoing TAO’s supply), and they auto-compound every ~72 minutes (one “tempo” = 360 blocks).

2. Two Revenue Streams and Why They’re Overlooked

In Bittensor, there are two completely separate ways to make money:

Revenue Stream 1: TAO Emissions via Taoflow

Since November 2025, Bittensor adopted the Taoflow model, a fundamental shift in how emissions are allocated.

Previously, emissions were calculated based on token price. This created a loophole: projects could artificially inflate their token price to capture emissions, build a “TAO treasury,” and slowly dump while still collecting rewards.

Taoflow fixed this by tracking net staking flow of TAO: staking inflows minus unstaking outflows. The mechanism runs in four steps:

After TAO’s first halving on Dec 14, 2025, block rewards dropped from 1 TAO to 0.5 TAO/block. Currently, ~3,600 TAO (~$960k at current prices) is distributed daily across 128 subnets. DCG estimates over $100M flows into the ecosystem annually.

Revenue Stream 2: Alpha Token PnL

This is the part most TAO holders aren’t tracking.

If a subnet performs well, Alpha’s price (denominated in TAO) rises. When you unstake, you receive more TAO than you initially put in. This is Alpha PnL = capital gains from holding a specific subnet’s token.

Taoflow creates a powerful flywheel effect:

  • Great product → more people stake TAO → positive net flow

  • Positive flow → more emissions → deeper liquidity pool

  • Deeper liquidity → lower slippage → attracts more capital

  • More capital → Alpha price rises → Alpha PnL increases for existing holders

The reverse is also true, and equally brutal. Sustained negative flow subnets → zero emissions → stakers continuously withdraw → death spiral.

3. Which Subnets Are Winning and Why?

Here’s a snapshot of leading subnets ranked by emission dominance and realized PnL (rPnL):

  • SN3| templar: Large-scale LLM pre-training | Emission Share: 30.39% | rPnL: $6.43M

  • SN4| Targon: AI inference market—hosting and serving AI models for real-time predictions | Emission Share: 10.39% | rPnL: $12.47M

  • SN68| METANOVA: AI drug discovery company developing therapies to reprogram body and mind | Emission Share: 5.95% | rPnL: $900k

  • SN81| grail: Verifiable LLM post-training processing | Emission Share: 4.8% | rPnL: $109k

  • SN75| Hippius: Decentralized storage and web infrastructure with IP management | Emission Share: 4.56% | rPnL: $4.48M

The top 10 subnets control ~56% of the total daily emissions.

Case Study: Chutes, The Textbook Example of Product-Market Fit (PMF)

Built by Rayon Labs, Chutes is a decentralized serverless AI inference market, a Web3 alternative to OpenAI API and AWS for model deployment.

What stands out about Chutes:

  • Processed 9.1T tokens since late 2024

  • 400k+ users (100k+ via API)

  • 85% cheaper AI model deployment costs vs. AWS

  • Models supported: DeepSeek, Mistral, LLaMA, and dozens more

  • Platform revenue auto-staked → buys back Alpha tokens → organic demand flywheel

During a spike in February 2026, Chutes attracted over 2,740 TAO in just 9 hours. Alpha tokens peaked at $99.94 (0.225 TAO), with an FDV of 2.05M TAO (~$518M at TAO’s peak).

Rayon Labs also operates SN56 (Gradients—model training) and SN19 (Nineteen—high-frequency inference), collectively commanding over 23% of total emissions at their peak.

Case Study: Templar (SN3), The Most Asymmetric Bet in the Subnet Ecosystem

On March 10, 2026, Templar (SN3) completed Covenant-72B, a 72B parameter model hailed as the largest decentralized pre-training run in history.

4. The Mechanics Behind It: Registration, Yuma Consensus, and Competitive Pressure

Registration, A Competitive Arena

Not everyone can open a subnet. Registration uses a dynamic burn pricing mechanism: the cost doubles with each new subnet registration and halves linearly over 28,800 blocks (~4 days) with no new registrations.

When all 128 slots are full, new subnets must replace the worst-performing existing subnet (measured by lowest EMA price). Newly registered subnets get a 4-month grace period before being eligible for deregistration. The network is expected to expand to 256 subnets in 2026.

Yuma Consensus, Automated Independent Auditing

Within each subnet, Yuma consensus transforms validators' subjective assessments into objective reward distribution:

  • Validators submit weight vectors, scoring each miner they evaluate

  • The blockchain calculates the stake-weighted median for each miner (kappa=0.5)

  • Weights above the median are clipped—to prevent collusion and over-evaluation

  • Validators use a commit-reveal mechanism—submitting sealed weights, revealed only after a set number of blocks—to prevent weight copying

  • Validators who identify good miners early and evaluate consistently build stronger bonded positions and earn larger dividend shares

The result: No subnet owner can unilaterally change who gets rewards. This is the fundamental difference between Bittensor and typical crypto “AI projects.”

5. Investment Framework: TAO = Index Fund, Alpha Staking = Startup Bet

Auditless Research summarized it perfectly: “TAO is effectively more of an options token—a call option that can be pointed at the emissions of whatever Alpha token looks undervalued.”

How to Read a Subnet Like a Balance Sheet

  • Emissions = Protocol subsidies - network revenue, similar to government grants or accelerator funding

  • Alpha PnL = Market cap signal - the market is pricing the subnet’s true value

  • Net staking flow = Revenue growth indicator - positive flow = “product sells,” negative flow = churning customers

  • Subnet owner = Founder quality - communication frequency, delivery speed, and product roadmap are things to track

  • Number of validators = Board quality - more independent validators = less chance of manipulated scoring

Institutional Signals Are Growing

This is no longer just a retail story:

  • DCG holds over 500k TAO (~2.4% of total supply)

  • Polychain Capital holds ~$200M worth of TAO

  • Grayscale GTAO Trust listed on NYSE on January 6, 2026

  • Stillcore Capital (co-founded by Jason Calacanis) launched a fund dedicated to subnet tokens

6. Conclusion: A Starter Framework

Bittensor has created a structure unique in crypto: 128 AI businesses competing for a shared ~3,600 TAO (~$960k) daily, where capital allocation is determined solely by staker behavior.

When evaluating a subnet, ask these five questions:

  • Product: What does this subnet offer? Is there real demand for it?

  • Flow: Is net staking flow positive or negative? What’s the 30-day trend?

  • Team: Is the subnet owner communicating and delivering consistently?

  • Flywheel: Does revenue create organic demand for Alpha, or is it pure speculation?

  • Exit: Is the liquidity pool deep enough to exit without significant slippage?

TAO gives you broad exposure to the entire ecosystem. Alpha staking gives you a concentrated bet on a specific “startup”—with all the accompanying upside potential and downside risk.

Related Questions

QWhat are the two primary ways to generate income within the Bittensor subnet ecosystem?

AThe two primary ways to generate income are: 1) TAO emissions (protocol subsidies based on staking inflow) and 2) Alpha token PnL (capital gains from the performance of the specific subnet).

QWhat is the Taoflow model and what fundamental change did it introduce in November 2025?

AThe Taoflow model is a fundamental shift in how emissions are distributed. It tracks the net staking flow of TAO (staking inflow minus unstaking outflow) into a subnet. Subnets with negative net flow receive zero emissions, forcing them to either perform or be eliminated.

QWhich subnet is highlighted as the clearest example of product-market fit (PMF) and what are its key metrics?

AChutes (SN64) is highlighted as the clearest example of PMF. Its key metrics include processing over 9.1 trillion tokens, having over 400,000 users (100,000+ via API), and offering AI model deployment at an 85% lower cost than AWS.

QHow does the investment exposure of holding TAO differ from staking for an Alpha token?

AHolding TAO provides broad, index fund-like exposure to the entire Bittensor network. Staking for an Alpha token is a concentrated investment in a specific 'startup' subnet, offering higher potential returns (100%+ APY) but also carrying significant risk.

QWhat mechanism ensures that no single subnet owner can unilaterally change who gets rewards, and how does it work?

AThe Yuma Consensus ensures this. Validators submit weight vectors rating miners, the blockchain calculates a median score, and weights above the median are clipped to prevent collusion and over-evaluation. A commit-reveal mechanism prevents copying weights, making the system an automated, independent audit.

Related Reads

Trump's 38th Call 'Deal is Imminent', Global Markets Experience TACO-style Surge

Former U.S. President Trump claimed for the 38th time that a "final agreement" is imminent, triggering a "TACO-style" surge in global stock markets. Major U.S. indices, as well as Japanese and South Korean markets, posted significant gains, with oil prices falling and gold rising. The market rally is attributed to renewed hopes for a ceasefire in the U.S.-Iran-Israel conflict, the latest U.S. CPI data showing inflation at a multi-year high but easing core pressures, and a significantly reduced expectation of a Federal Reserve rate hike this year. However, analysts warn of potential risks. The situation in the Middle East remains unresolved, with Iran denying any agreement. Several institutions have turned cautious, flagging overvaluation, speculative excess, and warning of a potential 6-7% correction in the S&P 500. Additionally, the massive SpaceX IPO, which has seen overwhelming retail demand, is testing market liquidity as investors may sell existing holdings to participate. Experts advise caution against market volatility driven by Trump's unpredictable statements.

Odaily星球日报3m ago

Trump's 38th Call 'Deal is Imminent', Global Markets Experience TACO-style Surge

Odaily星球日报3m ago

Retail Ecology Dwindles, ZKsync Bets on Bank Pilots for a Breakthrough

Amidst declining retail activity, ZKsync is pivoting to target institutional banking as its primary growth strategy. The article explores this shift, contrasting it with the competitive "survival of the fittest" narrative by highlighting a cooperative model inspired by naturalist Peter Kropotkin. ZKsync is developing infrastructure like its private, permissioned Prividium suite for banks (e.g., Deutsche Bank's use case via Memento), enabling private transactions with public verifiability via zero-knowledge proofs. This appeals to institutions needing privacy, compliance, and Ethereum-based settlement security, unlike fully private chains (e.g., JPMorgan's Kinaxis) or consortium models (e.g., R3 Corda). However, this strategic focus has coincided with a steep decline in its public DeFi ecosystem, evidenced by plunging TVL and the departure of major protocols like Aave due to low fees. The network's future now hinges on banking adoption, with upcoming pilots like the Cari Network involving regional banks holding over $600 billion in deposits. A significant challenge is balancing this institutional focus with ZKsync's decentralized governance. Banks must operate on a network where rules and fees (denominated in the volatile ZK token) can be changed via community vote, and where a Security Council holds emergency control—a stark contrast to the predictable, contract-bound environments of traditional finance. The coming 18 months will test whether ZKsync can successfully onboard traditional banks onto a dynamically governed public chain or if institutions will ultimately revert to proprietary solutions.

Foresight News15m ago

Retail Ecology Dwindles, ZKsync Bets on Bank Pilots for a Breakthrough

Foresight News15m ago

The Recursive AI Anthropic Warned About: Tian Yuandong's New Company Has Just Taken the "First Step"

Anthropic recently highlighted the rapid progress toward "recursive self-improvement," where AI systems autonomously design and train their successors. In response, Recursive Superintelligence, a new company co-founded by former Meta researcher Tian Yuan Dong, has publicly demonstrated its first step toward automating AI research. The company released a system designed to autonomously execute the full AI research cycle: generating ideas, implementing code, running experiments, and learning from results. It validated this approach by achieving state-of-the-art results on three diverse benchmarks: 1. **NanoChat Autoresearch:** Optimizing a small language model's validation loss under a fixed 5-minute GPU budget, improving upon the community's best result. 2. **NanoGPT Speedrun:** Reducing the time to train a GPT model to a specific loss on 8 H100 GPUs from 79.7 seconds to 77.5 seconds, beating a highly optimized, human-driven community effort. 3. **SOL-ExecBench:** Improving the overall score on NVIDIA's suite of 235 GPU kernel optimization tasks by 18%, closing the gap to the hardware limit. The system discovered novel optimizations in this highly specialized domain without direct human expertise. Recursive's system operates as a general framework, capable of parallel exploration and cross-task knowledge transfer while incorporating safeguards against reward hacking. The company, backed by $650M in funding and a star-studded team including Richard Socher and Alexey Dosovitskiy, aims to create AI that recursively enhances its own research capabilities. This development represents an early but concrete move toward a new paradigm where AI accelerates its own advancement. It occurs alongside Anthropic's warnings about the need for industry coordination and potential pauses when recursive self-improvement thresholds are reached, highlighting the dual trajectory of rapid technical progress and growing calls for careful stewardship.

marsbit23m ago

The Recursive AI Anthropic Warned About: Tian Yuandong's New Company Has Just Taken the "First Step"

marsbit23m ago

The Gold Buy-on-the-Dip Guide: Watch Interest Rates, Not Just War

"Gold Buying Guide: Focus on Interest Rates, Not Just War" Four months ago, gold buyers likely didn't anticipate buying at a peak that even a war couldn't sustain. After hitting a record high of $5,596 on January 29, gold entered a bear market just 91 days later, its fastest decline since 2008. A key trigger was the Fed's hawkish shift, highlighting that monetary policy, not geopolitics, is the primary driver. The article argues that the traditional "buy gold in turmoil" script has changed. While the US-Iran conflict initially boosted prices, the sustained rally in oil prices heightened inflation fears, forcing central banks to maintain or consider tighter policy. Since gold yields no interest, higher rates increase its opportunity cost, eroding its appeal. This dynamic was evident when gold fell sharply on May 18 despite positive peace talks, as lower oil prices eased inflation and thus rate hike pressures. The recent sell-off is also part of a broader market deleveraging. Correlations between gold, Nasdaq, and Bitcoin spiked as leveraged investors sold liquid assets to cover losses, creating a synchronized downturn. Historically, gold bottoms align with policy shifts, not conflict resolutions. The 2008 and 2022 bear markets ended with shifts to extreme easing and peak inflation expectations, respectively. For potential buyers, the author suggests monitoring three signals: 1) Peak interest rate hike expectations, 2) Reopening of the Strait of Hormuz (to ease oil/inflation pressure), and 3) A return to net inflows for Gold ETFs, indicating the end of forced selling. While predicting the exact bottom is impossible, the author's personal strategy involves scaling into a position across price levels like $4000, $3700, and $3500, committing no more than 30% of the intended total allocation initially, and adding the remainder only if key signals emerge. The core conclusion: In turbulent times, watching interest rates is more crucial than watching wars.

marsbit29m ago

The Gold Buy-on-the-Dip Guide: Watch Interest Rates, Not Just War

marsbit29m ago

Recent On-Chain Review: No Clear Narrative Under U.S. Stock Market Pressure, Just Hype

This article analyzes the current state of the Solana meme coin and community token ecosystem, highlighting a market caught between two dominant forces: attention-based PvP and a gradual return to community-centric projects. The first part explores the "Attention PvP" dynamic, where success is driven by celebrity endorsements, viral events, and speed. Examples include $JOTCHUA, which surged after its meme creator's social media activity, and $WORLDCUP, which outperformed a similar Base chain project ($PITCH) largely due to influencer support. The recent "pump.fun GO" feature, allowing bounty tasks for token promotion, is critiqued for fostering sensationalist and often negative stunts—like people getting token tickers tattooed on their bodies for rewards—reminiscent of old internet shock content. In contrast, the article points to a resurgence of organic, community-driven tokens that survive market volatility through strong holder bases and shared ideology, not just hype. Influencer Ansem is cited, arguing that durable meme coins rely on communities willing to endure losses and promote their core message daily. Examples given are older tokens like $neet (anti-work ethos), $troll, $buttcoin, and $triplet, which have maintained relative price stability. A prime example of this community-build model is the new project $KINS, the token for the browser-based MMORPG Kintara. Its success stems not from advanced graphics but from consistently delivering updates, fostering player trust, and creating genuine engagement (e.g., in-game economies, events, property auctions). It has attracted a growing player base and even notable KOLs as participants, demonstrating that sustainable growth can come from building trust rather than orchestrating pumps. The article concludes by questioning whether the market is ultimately a game of mutual trust or mutual deception, expressing hope that such reflection might lead to a healthier ecosystem.

marsbit29m ago

Recent On-Chain Review: No Clear Narrative Under U.S. Stock Market Pressure, Just Hype

marsbit29m ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of S (S) are presented below.

活动图片

Top Questions

Hot Categoriesright-arrow

Hot Tagsright-arrow