funny how so many identical posts are coming up about this. some people are desperate to pull down tao - probably stooges of the tech oligarchs. People should ask what level of subsidies were and are still in play for the centralised AI models.

Bittensor (TAO) current price is $275, market cap is $2.6 billion, but the subsidy income ratio of the core subnet Chutes is as high as 22:1 to 40:1. After removing the subsidies, the cost is more than 1.6 times that of centralized solutions; the valuation multiple of 175-400 times far exceeds industry standards, and what the market is trading is a narrative of scarcity rather than real fundamentals. This article is sourced from a piece by Pine Analytics (The Bear Case for Bittensor (TAO)), edited, translated, and organized by Dongqu. (Background: Balancer founder announced "dissolve the team, retain the protocol," after a hack, TVL remains at $158 million, subsequently taken over by DAO) (Background: What is Terafab? Musk calls for a global chip shortage demand of 2%. How to build a factory "larger than TSMC"?) TAO's current price is about $275, with a market cap of $2.6 billion and a fully diluted valuation of $5.8 billion. The project has received backing from Grayscale (application for ETF listing on the NYSE submitted in December 2025), and has also received public recognition from NVIDIA CEO Jensen Huang. At the same time, the narrative of token supply is very attractive: total cap of 21 million tokens, using a Bitcoin-like halving mechanism. After the first halving in December 2025, the daily issuance will drop from 7,200 tokens to 3,600 tokens. Within one year, the number of subnets increased from 32 to 128, and Templar's Covenant-72B training has also proven that decentralized computing power can run large language models with competitive benchmarks. This report does not deny the above facts. What we want to explore is whether the economic model of this network can generate real external income to support the current valuation scale and how its competitiveness is when competing against centralized service providers and self-hosted computing power. Bittensor (TAO) token issuance distribution ratio. There are four types of participants in Bittensor: · Subnet owners build professional AI markets, receiving 18% of TAO issuance rewards from the subnet; · Miners execute AI tasks (inference, training, data processing), receiving 41%, totaling about 1,476 TAO per day, with an annualized value of about $14.8 million; · Validators score the output of miners, receiving 41%; · Stakers put TAO into subnet liquidity pools in exchange for subnet-specific tokens. Under the Taoflow model, a subnet's reward share is determined by the net inflow of staked TAO, with no reward if the net inflow is negative. The top ten subnets control about 56% of the total network issuance. TAO is the universal token for the entire network: miners register, validators stake, subnet tokens are purchased, and service payments must all use TAO. Theoretically, subnet activities will generate structural demand for the underlying token. Bittensor subnet Chutes (SN64) and centralized service provider LLaMA 70B model inference cost comparison analysis. Bittensor's supply side is highly transparent: 3,600 TAO are allocated daily according to a program, with halving rules hard-coded; staking rate (approximately 70%), distribution ratio, and flow data are all on-chain. But the demand side is completely opaque. There is no unified dashboard to track external income by subnet, and the actual calls for AI services (inference, computation, training) occur off-chain and are not recorded on the blockchain. Investors can only infer demand through indirect indicators such as staked flows, subnet token prices, and self-reported data from the project party. This opacity is structural, not a temporary phenomenon. The blockchain only records token circulation, not API calls. Below is the most complete demand-side profile as of March 2026. Chutes occupies 14.4% of the total network issuance, the highest among all subnets. Developed by Rayon Labs, it provides open-source model serverless inference services, priced 85% lower than AWS and 10%-50% lower than Together AI. Its usage data is unparalleled in the ecosystem: over 400,000 users (over 100,000 API users), daily request volume exceeds 5 million times, cumulatively processing 9.1 trillion tokens, with the three-day average token generation increasing from 6.6 billion to 10.1 billion. It is also a top inference service provider on OpenRouter, with some models outperforming centralized competitors. However, this low price does not come from operational efficiency but from subsidies. Based on a 14.4% share, Chutes receives about 518 TAO daily, with an annualized value of about $52 million. Its external annual income is only about $1.3 million to $2.4 million (the higher value is self-reported by the team and not independently audited). The subsidy ratio for this subnet is about 22:1 to 40:1. For every $1 paid by users, the network needs to release $22-$40 of TAO through inflation to subsidize. If subsidies are removed, based on its daily processing of about 10.1 billion tokens, the cost price is about $1.41 per million tokens. The current centralized market prices are: · Together.ai's LLaMA 3.3 70B Turbo is about $0.88 per million tokens; · DeepSeek V3 is about $0.40-$0.80; · Smaller models can go as low as $0.18. This means that without subsidies, Chutes' price would be 1.6–3.5 times higher than centralized solutions. The so-called 85% cost advantage is fully reversed, and its low price is fundamentally a purchase order by TAO holders through inflation, not structural efficiency brought about by decentralization. When the next halving comes (expected at the end of 2026 or 2027), either the price will double, or miners will exit, or the gap between subsidies and income will further widen. Some may draw parallels to early internet subsidies for customer acquisition, but Uber, DoorDash, and AWS established switching costs during their subsidy periods: proprietary platforms, driver networks, and enterprise ecosystems. However, Bittensor subnets have no barriers: models are open-source, interfaces are standardized, and users can switch service providers at zero cost. Once subsidies recede, there are no locking mechanisms to retain users. Rayon Labs also operates SN56 and SN19, collectively controlling about 23.7% of the total network issuance, with no external income disclosed. A single team almost controls one-fourth of the network's incentive distribution. Targon (SN4) is the highest revenue subnet, operated by Manifold Labs, providing confidential GPU computing services to enterprises, with an estimated annual revenue of about $10.4 million, corresponding to a valuation of $48 million, with a market-to-sales ratio of about 4.6 times, the most solid valuation in the ecosystem. But $10.4 million is just predicted data quoted from multiple reports, not audited figures. Templar (SN3) has completed the Covenant-72B training, with a market cap of $98 million, but external income is zero. Training API and enterprise sales are still progressing, with no paid product launched yet. The remaining 120+ subnets either have no public income or are still in the early product stages, mainly relying on token issuance subsidies.