Originally published at mrnasdog.com/research/tao/full by MrNasdog.
This is a MrNasdog Pressure Framework analysis of Bittensor (TAO) on Metric 1 (sell pressure) and Metric 2 (buy pressure). Narrative (Metric 3) is covered separately. The short version: TAO has the highest annual inflation in our coverage today (~27% of circulating), but it also has the only programmed structural catalyst in the lineup — a BTC-style halving in ~10 months that automatically halves the emission, with no governance vote needed.
The setup
Bittensor (TAO) is an AI-focused, Substrate-based proof-of-stake network. It launched with no ICO, no pre-mine, and no team allocation — every TAO in existence was minted from block rewards. The max supply is 21M (BTC-style). Block reward starts at 1 TAO/block and halves every 10.5M blocks (~4 years), so the emission curve mirrors Bitcoin's.
Live parameters, origin-first from the Subtensor mainnet RPC (entrypoint-finney.opentensor.ai):
-
Current block: 8,249,997 (read via
chain_getHeader) - Halvings completed: 0 (we're ~78.6% through the first cycle)
- Current emission: 1.0 TAO/block → ~2.63M TAO/year
- Blocks to first halving: 2,250,003 → ~10 months (around Q1–Q2 2027)
- Active subnets: 60+ (Dynamic TAO active since Feb 2025)
- Circulating supply (CoinGecko cross-check): ~9.6M TAO
- Max supply: 21M TAO (fixed, BTC-style)
- Price: ~$280 → market cap ~$2.68B · FDV (at 21M) ~$5.87B
The big number: 2.63M new TAO per year against 9.6M circulating ≈ ~27% annualized dilution today. That's the highest inflation rate by far in our analyzed lineup. After the first halving it drops to ~1.31M/yr against ~11.8M circulating ≈ ~11%/yr. After the second halving (~2031), ~5%/yr. The curve looks exactly like BTC's, in years 3–4 of the same shape.
The 2025 Dynamic TAO (DTAO) upgrade changed how emission is distributed across subnets (price-weighted alpha tokens per subnet, root TAO claim mechanics) but did not change the headline 1-TAO-per-block cap or the halving schedule.
The sell ledger
What the design predictably puts on the market.
| # | Source | Tag | Value |
|---|---|---|---|
| 1 | Protocol inflation | Tag A | ~2.63M TAO / yr (~27% of circulating) |
| 2 | Vesting unlocks (still-locked allocations) | — | 0 (no pre-mine, no team allocation, no ICO) |
| 3 | Team / DAO / identified-group holdings | Tag B | OpenTensor Foundation + key founder/operator wallets — TBD pending enumeration |
| 4 | Bankruptcy estate distributions | — | 0 |
The headline is inflation. At 1 TAO/block × ~7,200 blocks/day, TAO mints ~7,200 new tokens daily, ~2.63M/year. At today's $280, that's ~$2 million / day of structural sell-pressure cost — paid to validators, miners, and (since DTAO) subnet stakers. The portion they sell to cover operating costs is the actual market sell pressure.
Vesting: zero. This matters a lot for Bittensor: there is no founder cliff, no investor cliff, no team unlock. Everything was earned through mining/staking participation. So source #2 contributes nothing today, and won't tomorrow either.
Tag B is the OpenTensor Foundation + key validator entities. OpenTensor Foundation (the steward of the protocol) accumulates TAO via subnet participation and stewardship rewards. Founder (Const / Jacob Steeves) and operator entities (Yuma Group, RAO Labs, etc.) are visible on the chain but require subnet-by-subnet enumeration to quantify. Flagged as TBD here — same caveat we used for ONDO's Foundation Safe and NEAR's lockup contracts.
Bankruptcy estate: zero. No FTX-style estate distributing TAO.
The buy ledger
What the design predictably takes off the market.
| # | Source | Value |
|---|---|---|
| 1 | Revenue-backed buyback | 0 — no structural buyback contract |
| 2 | Burn mechanism (subnet UID registration "recycle") | Tag A, small — TAO burned to register new neurons on subnets |
| 3 | Locked allocations | — context only (substantial TAO staked, but unbond ~7 days = functionally liquid) |
| 4 | Protocol-level demand (subnet activity) | Tag A — subnet registration + AI inference payments, real but specialized |
There is no protocol buyback. TAO has no revenue-funded contract buying back from market.
The structural burn is subnet-registration "recycle." When a participant registers a new UID on a subnet, the registration cost (denominated in TAO) is burned — it goes out of circulation. The cost is dynamic (rises with demand to register slots). At today's subnet activity it's a real but modest burn — single-digit thousands of TAO per day across all subnets. The post-DTAO model also introduces emission rebalancing that effectively removes inflated emissions from underperforming subnets; these are technical mechanics but the headline burn line is still subnet registration.
Protocol-level demand exists. TAO is needed to participate in subnets (register UIDs, claim alpha emissions), and an emerging market for AI inference payments uses TAO. But this demand is narrower and more specialized than (for example) BSC gas demand — it's primarily participation-driven, not user-facing utility.
Net position
Combine the ledgers and the picture today is brutal on the supply side:
- Sell, Tag A: ~2.63M TAO / yr (inflation)
- Buy, Tag A: modest (subnet registration burn + protocol demand, both small relative to inflation)
- Net structural dilution: roughly ~25%+ / year at current emission
This is the same shape as Ondo and NEAR — scheduled supply up, structural buy minimal — but larger in magnitude than either. TAO has the highest annual inflation in the lineup right now.
But the catalyst makes TAO different from anything else we cover.
The halving — the only programmed catalyst in the lineup
Every other coin we've analyzed depends on a governance decision to flip its buy ledger: Ondo needs a DAO fee-switch vote, NEAR needs the protocol_reward_rate restored, HYPE's buyback is structural but ratified by validator vote. TAO's catalyst is automatic and code-level — at block 10.5M (in ~10 months), the block reward drops from 1 TAO to 0.5 TAO. No vote, no governance, no discretion. It happens.
What that means for the framework:
- Pre-halving (now → Q1 2027): ~2.63M TAO/yr emission, ~27% inflation
- Cycle 2 (post-H1, ~2027–2031): ~1.31M TAO/yr, ~11% inflation
- Cycle 3 (post-H2, ~2031–2035): ~0.66M TAO/yr, ~5% inflation
- Cycle 4 (post-H3, ~2035–2039): ~0.33M TAO/yr, ~2.5% inflation
- By the 4th halving (~2039): TAO is roughly at BTC's current inflation profile
This is the BTC playbook, 12 years offset. The framework gives TAO an unfavorable read today on Metric 1 + Metric 2, but the structural catalyst is real, scheduled, and the most predictable buy-side improvement available across the coins we cover.
What to watch
-
The first halving event — block 10.5M, ~Q1 2027. Confirm via the Subtensor RPC's
chain_getHeadernear that block. - DTAO subnet emission distribution — which subnets are absorbing alpha emissions vs. swapping to TAO; this affects validator sell-pressure dynamics even before the halving.
- Subnet registration burn rate — rises with demand to register, falls when registration slots cool. A real-time meter on chain activity.
- OpenTensor Foundation stake + key operator wallets — once enumerated, sets the Tag B size precisely.
MrNasdog Pressure Framework analysis of TAO, Metrics 1 & 2. Data + explanation only. Not financial advice. Numbers as of May 2026.
Data note: All chain-level numbers are origin-first from the Subtensor mainnet RPC (entrypoint-finney.opentensor.ai). Block height, emission rate, and halving math read via chain_getHeader and the public docs at docs.bittensor.com/learn/emissions. Halving timing per the December 2025 official announcement on docs.bittensor.com/learn/announcements. Price and circulating supply cross-checked via CoinGecko. OpenTensor Foundation + key operator balances flagged as TBD pending subnet-aware enumeration.
Top comments (0)