Ora Protocol

What it does

Ora Protocol was a verifiable oracle that brought AI inferenceInferenceRunning a trained AI model to produce an answer. Inference is what happens when you type a prompt into ChatGPT and get a response. The model takes your input, computes a best guess, and returns it.Like asking an expert for their opinion. The training was the decades they spent becoming an expert. The inference is the 30 seconds it takes them to answer your specific question.Inference in the glossary → on-chain. Smart contracts could call it for a model output and get back a result carrying a cryptographic guarantee that it had been computed honestly. That description is now historical.

Three things happened, and none of them was announced.

The oracle stopped serving. On Ethereum, the Onchain AI Oracle contract at 0x0A0f4321214BB6C7811dD8a71cF587bdaF03f0A0 counts every request it has ever received. That counter reached 4,615 on 26 September 2025 and has not moved since. The last request answered in ordinary time was number 4,610, on 29 April 2025.

The documentation was deleted. The site at docs.ora.io served content as recently as 22 April 2026 and now returns no DNS record at all. It took the tokenomics page, the node-operator guide and the whitepaper link with it. Search engines still index those pages, so anyone following a link from a paper or a tutorial lands on nothing.

The name was reused. Between 5 May and 7 June 2026, ora.io was repointed at an AI-managed prediction-market fund product. It kept the Ora branding and expanded the acronym to Optimized Return Agents.

This is a pivot, not a hijack

An apex domain that stops resolving while a new site appears on the www subdomain is the shape of a lapsed registration or a registrar takeover, so that was worth ruling out before anything else. That’s ruled out four ways:

• The domain never lapsed. WHOIS shows ora.io active at GoDaddy with registry expiry in May 2027, on the same Amazon Route 53 nameservers it has always used.
• The GitHub org is the same org. The ora-io organisation still holds the opml, OAO and awesome-ora repositories, with their history intact, and now describes itself as Optimized Return Agents.
• The npm publisher is the same account. The new agent SDK package is published by hyper-oracle, the account Ora used under its original HyperOracle name.
• The X account is the same account. The verified handle joined in January 2022 and carries the same follower base under the new bio.

So the team that built the oracle is the team running the funds. What did not survive the move is the token.

Value proposition

The engineering was worth taking seriously, which is why the ending matters. opML (optimistic machine learningMLMachine Learning. The branch of AI where systems learn patterns from data instead of being explicitly programmed with rules. Modern AI (LLMs, image generation, recommendation systems) is almost entirely machine learning.Like teaching a child to recognise dogs by showing them thousands of pictures of dogs, instead of writing down a precise rulebook for what makes a dog. The child learns the pattern from examples rather than from instructions.ML in the glossary →) ran inference off-chain and posted the result on-chain as correct unless challenged. A challenge triggered a bisection game that narrowed the disagreement to a single computational step, which a Fraud Proof Virtual Machine then settled on-chain. Because verification only ran on dispute, it cost far less than zkMLZKMLZero-Knowledge Machine Learning. A cryptographic technique that lets a prover convince a verifier that a specific model produced a specific output, without revealing the model weights or the input.A sealed-bid auction with a referee. You don't see the bids and you don't see the model that picked the winner, but the referee hands you a tiny receipt that proves the winner was selected by the agreed rules. ZKML is the maths that produces that receipt.ZKML in the glossary →, which proves every computation whether anyone doubts it or not.

The team also authored three Ethereum standards: ERC-7641 for revenue-sharing tokens, ERC-7007 for verifiable AI-generated content, and ERC-6150 for hierarchical NFTsNFTNon-Fungible Token. A unique blockchain-tracked asset where each token is distinguishable from every other. Where regular tokens are interchangeable, NFTs represent unique items like art, collectibles, in-game assets, or domain names.Like the difference between a $20 note and a signed first-edition novel. The notes are interchangeable, any $20 buys the same thing as any other. The book is one of a kind, and its value depends entirely on which specific book it is.NFT in the glossary →. Those are in the public record permanently. They are the part of this project that outlives it.

What the chain says about usage

Usage figures of more than 180,000 inference calls, over 500 nodes running the Tora client and 20-plus projects building on the oracle circulated around it, but we couldn’t trace them to any Ora page that still opens, so they can’t be re-checked. The contract can be read directly instead, and it tells a smaller story.

Reading, Ethereum mainnet	Value
Lifetime requests received	4,615
Last request received	26 September 2025
Last request answered on time	29 April 2025
Requests never answered	4,611 to 4,615
Only result event since	5 May 2026, settling request 4,606

Two details there carry most of the weight. The single result posted in May 2026 answered request 4,606, an older one, so it was a backfill rather than a return to service. And every request from April 2025 onwards came from 0x61423153f111BCFB28dd264aBA8d9b5C452228D2, which Ora’s own oracle repository names as its SimplePrompt example contract. In its final months the oracle’s only caller was Ora’s own demo.

Elsewhere the picture is the same. The last activity readable on Base was May 2025, on Arbitrum August 2025, and on Linea March 2025.

What replaced it

The replacement product listed AI-run funds that trade prediction markets, and it was small: six funds with a combined AUM in the low hundreds of dollars, and a headline Ora World Cup leaderboard showing no ranked entries. The team was shipping, with an agent SDK published in June 2026 that submits orders to Polymarket in USDC.

Then that went dark too. On 18 September 2026 ora.io answers NOERROR with no A record at all, on both Cloudflare and Google, so the domain is still registered and still delegated to the same Route 53 nameservers and serves nothing. The docs subdomain went first, the oracle stopped emitting a year ago, and now the product that took the name is unreachable as well.

The ORA token has no part in it. The site’s rendered page and its JavaScript bundles contain no occurrence of the token’s contract address, its ticker, its tokenomics, opML, the oracle, the model-offering programme or stakingStakingLocking up a cryptocurrency to help secure a blockchain network, usually in exchange for rewards. The locked tokens act as a security deposit that can be taken away if the staker misbehaves.Like putting down a large rental deposit for an apartment. You get the money back if you behave, you earn interest while it's locked, and the landlord takes it if you trash the place.Staking in the glossary →.

Tokenomics

The token contract is the strongest thing left, and it’s stronger than a fixed supply usually implies. Reading the deployed contract directly: there is no owner, no role admin, no minter and no pause, every corresponding call reverts, and both proxy storage slots read zero, so the logic cannot be replaced. The verified source carries no mint function beyond four constructor calls, closed by an assertion that total supply equals the cap. Supply can fall through burns and cannot rise. Contract address: 0x33333333FEde34409Fb7f67c6585047E1F653333.

Set against that, the disclosure record is poor and has now been made permanent. A single on-chain transfer of exactly 110,000,000 ORA on 16 December 2025 moved circulating supply from roughly 17% to roughly half, with no published schedule and no announcement. Holders could not have anticipated it, and cannot anticipate what remains. Around half of total supply still has no public allocation, and $23M was raised from institutional investors whose token terms were never disclosed.

On-chain concentration is heavy. The top ten Ethereum holders control 93.6% of supply raw, and 80.4% after excluding the bridge adapter as neutral infrastructure. Around 77% resolved to insiders when the funding trace last ran, once two team multisigs and a Safe-funded wallet were attributed. Ora described the launch as a Decentralized AI Community Offering modelled on Vitalik Buterin’s DAICO idea, which describes the initial float rather than the supply sitting behind it.

Every documented use for the token has now lapsed:

• Node staking concluded before the oracle stopped.
• Inference fees have no inference to price.
• Model-offering participation and agent creation ran on a site that no longer exists.
• Governance was listed as a core token utility in Ora’s own documentation and never had a venue to exercise it.

That last one deserves its own line. Ora’s ORA Coin page told holders they would shape the future of the protocol through decentralised governance. No Snapshot space was ever registered under any plausible identifier, and no on-chain governance contract was ever deployed. When the future of the protocol was actually decided, holders were not asked, and the page carrying the promise was taken down rather than honoured.

How to participate

There is no live participation route. Node operation ended with the inference network, building on the oracle means integrating with something that does not answer, and the agent-creation product went down with its website.

If you hold ORA, the practical position is a token with an immutable supply, a heavily concentrated holder base, thin decentralised liquidity and no product behind it. If you came here for the technology, the standards and the opML papers are still published and still worth reading.

Honest assessment

Freedom Score: 25/100

Infrastructure Decentralisation: 3/20. The network is not answering. The Ethereum request counter has been static since September 2025, five requests from that period were never served, and the closing traffic came from Ora’s own demo contract. The 500-node figure was self-reported and cannot be re-checked, because the node-operator guide lived on the deleted documentation domain.

Governance Decentralisation: 2/20. Governance was published as a core token utility and no venue was ever built. Querying the Snapshot hub for the plausible Ora identifiers returns nothing, and no on-chain governance contract is deployed. The decision to retire the product and reuse the name was taken without a holder vote, notice or wind-down statement.

Token Distribution Fairness: 5/15. The great majority of supply sits with insiders on-chain, as the tokenomics section sets out. The community framing described the initial float. Roughly half of supply entered circulation in one unannounced transfer, and the party that could have documented it has deleted its documentation.

Censorship Resistance: 7/15. Re-scoped to the token layer, which is the part still running, and it’s clean there. No owner, no roles, no pause, no blacklist and no upgrade path, so no balance can be frozen or seized. The AnyTrust argument that previously carried this dimension no longer applies, because there is no live inference to censor.

Data Sovereignty: 3/15. The verifiable-inference provenance that justified this score is no longer produced. The 4,615 historical requests remain permanently readable, which is worth something, and nothing new is being written.

Open Source Transparency: 5/15. The code and the standards hold up: 46 public repositories, an MIT-licensed opml repo, three authored ERCs, and a Salus Security audit of the staking contract dated 13 July 2024 reporting 2 high, 3 medium and 6 low findings with 11 of 13 resolved. The publication surface around them has collapsed. The documentation site is deleted, the blog has not posted since January 2025, the foundation subdomain no longer resolves, no audit ever covered the opML or oracle core, and the pivot was announced nowhere the project owns.

Returns Score: 16/100

Token Utility: 2/20. Staking concluded, inference fees have no inference, the model-offering and agent products went down with their website, and governance never had a venue. Axon liquidity mining is the one mechanism not shown to be dead, and its documentation has been deleted. The successor product does not mention the token anywhere.

Value Accrual: 1/20. There is no fee to accrue. The oracle’s model, callback and network fees needed inference that stopped. ERC-7641 revenue sharing applied to model-offering tokens rather than to ORA, and no buy-back or burnBurnPermanently removing tokens from circulation by sending them to an address that no one controls. Burns reduce total supply, which (all else equal) makes each remaining token worth more of the network's value.Like a company buying back its own shares and shredding them. The company's total value stays the same, but each remaining share now represents a slightly bigger slice of that value.Burn in the glossary → was ever implemented. Whatever the funds product earns, it earns in USDC and routes none of it here.

Supply Dynamics: 9/20. Held up almost entirely by the contract. No mint, no admin, no proxy, and a supply that can only fall. The deductions are for the undisclosed half and the unannounced December 2025 release, which are disclosure failures rather than code ones.

Revenue Sustainability: 1/25. The evidence that carried this dimension described a network that has stopped. The Ethereum lifetime counter stopped at 4,615, and the last traffic was self-generated. The successor funds are live at a combined AUM in the low hundreds of dollars.

LiquidityLiquidityHow easily a token can be bought or sold without moving the price. High liquidity means you can enter or exit large positions quickly at the quoted price. Low liquidity means even small trades can swing the market.Like the difference between selling a house and selling a share of Apple stock. The house might be worth more on paper, but finding a buyer at that price takes weeks. The Apple share converts to cash in one click.Liquidity in the glossary → & Access: 3/15. Four tickers on four decentralised venues, with Uniswap V3 on Ethereum carrying about 84% of the flow. No centralised venue quotes ORA. Median daily turnover sits close to zero, under a hundredth of a percent of market cap in early October 2026, and the token now has no product behind it to attract a listing.

Quadrant: D (Low Freedom, Low Returns)

Ora leaves Quadrant B and lands in D. The technical work was substantive and the standards contributions are permanent, but a protocol that has stopped serving cannot score on infrastructure, usage or data, and a token its own team has walked away from cannot score on utility or accrual.

Key risks

• The protocol is dormant. The oracle has taken no request since September 2025 and answered none since April 2025.
• Five requests were never answered. Contracts that called the oracle in August and September 2025 paid and are still waiting. Unanswered is worse than idle.
• The team has moved on without saying so. No blog post, documentation update or announcement covers the retirement of the oracle or the future of the token.
• The documentation is gone, permanently. Tokenomics, node guide and whitepaper links all dead-end, so nothing about the token’s design can now be checked against a live first-party source.
• About half of supply is undisclosed. No allocation, no vesting schedule, and a 110M release that arrived unannounced.
• The great majority of supply is insider-controlled on-chain, against thin decentralised-only liquidity.
• No audit covers the oracle core. The published Salus audit covers the staking contract only.
• The successor product is tiny. Six funds at a combined AUM in the low hundreds of dollars is not evidence of a team that traded up.

Score change log

Date	Score	Change	Reason
2026-10-07	Data	N/A	Turnover restated from around a tenth of a percent to under a hundredth (7-day median 0.001%, early October). Scores unchanged.
2026-09-24	Data	N/A	Turnover wording moved from under a tenth of a percent to around a tenth, matching the 0.112% 7-day median of 23 Sep, and the ATH decline rounded to ~99.9%. Scores unchanged.
2026-09-23	Data	N/A	Integrity records re-cut: 4 claims now quote Ora’s own pages verbatim and 6 are our own measurements. Scores unchanged.
2026-09-18	Data	N/A	ora.io now serves no A record at all on two resolvers, so the prediction-market product that took the name is unreachable as well. Wired the DNS, Snapshot and oracle-liveness checks to weekly refreshers; scores unchanged.
2026-09-08	Both	Freedom 47 → 25 (F), Returns 43 → 16 (F)	The oracle is dormant and the project pivoted. Ethereum’s request counter stopped at 4,615 on 26 Sep 2025 with requests 4,611 to 4,615 unanswered, and the only later result backfilled an older request. Status set to dormant.
2026-09-08	Data	N/A	Pivot established rather than assumed. Lapse and takeover ruled out via WHOIS, the ora-io GitHub org, the hyper-oracle npm account and the X account; transition dated to 5 May to 7 Jun 2026 by capture digests. The successor product references the token nowhere.
2026-09-08	Data	N/A	Governance verdict restored to inflated after being withdrawn earlier the same day. The withdrawal generalised one uncaptured page to the whole host; the archived ORA Coin page carries the governance claim verbatim. Claim sources also recovered for supply and distribution.
2026-09-06	Data	N/A	Research JSON resynced to the frontmatter Returns breakdown. Supply Dynamics read 10 in the JSON after the 2026-09-05 move to 11 never propagated. No published score changed.
2026-09-05	Both	Supply Dynamics 10 → 11 (Returns 42 → 43), Token Distribution Fairness 7 → 6 (Freedom 48 → 47)	Resolves the 15 August deferral: the dilution overhang halved when circulating was corrected to ~50%, but the same 110M arrived with no published schedule, so disclosure reads worse. Both grades held at D.
2026-08-16	Data	N/A	Corrected “No audits found”: Salus Security audited the staking contract (13 July 2024), 2 high / 3 medium / 6 low, 11 of 13 resolved, though no published audit covers the opML core. Open Source Transparency 10/15 held pending monthly review.
2026-08-15	Data	N/A	Circulating supply corrected from 16.9% to 49.89% (~166.3M) and FDV/MCap from 5.9x to 2.0x, after an on-chain transfer of exactly 110,000,000 ORA on 16 Dec 2025 (Ethereum block 24025314). Supply Dynamics 10/20 and Token Distribution Fairness 7/15 held pending monthly review; weighing in docs/quarterly-review-backlog.md.
2026-08-13	Data	N/A	ORA has no citable published source for the distribution claim: docs.ora.io no longer resolves and ora.io has pivoted to an agent-funds product with /token, /tokenomics, /docs and /blog all 404. Recorded as unsourced rather than cited to third-party press.
2026-08-10	Data	N/A	WS3 holder-source + funding-trace: authored token_distribution_fairness (inflated, onchain), confirming the 7/15 score. Ethereum top-10 ex-infra 81.4% of supply, ~77% insider: two team/treasury Safes plus a 26.3% EOA the trace resolved to Safe-funded. The “community-first TGE” describes the float, not the concentrated supply. No score change.
2026-06-26	Data	N/A	DefiLlama LlamaAI harvest cross-check, verified independently (api.llama.fi + CoinGecko, coingeckoId ora-coin). Published figures held up: 16.9% circulating (~56.3M of 333.3M), sub-$1M market cap ($264K), $23M total funding. Corrected the FDV/MCap ratio (5.2x → 5.9x; 16.9% circulating mathematically implies 5.9x). Re-verified ATH decline at ~99.9% (CoinGecko -99.91% from $5.37, TGE day 19 Nov 2024). Two DefiLlama warehouse bugs noted but they do not touch our data: the broken coingecko:ora-protocol price map is a slug we do not use (ours, ora-coin, resolves), and the double-logged $20M Series A (summed to $43M under ora+ora-protocol) is already deduped to the correct $23M in our copy. No score change.
2026-03-06	Both	N/A	Initial publish. Freedom 48/100, Returns 42/100.