Llama was an onchain governanc…me of the firm that built it. The firm spent 2021–2022 as one of the better-knownIn 2021–2022 the firm worked as a DAO service providers, writing and shepherding prop…very action to a token vote. The firm is gone. This page is not about that – the shutdownRiver Protocol acquired the firm in November 2024. Its end, and the general problem it b… is one of a measured cohort. What is documented here iThis page covers what the framework did, who actually randeployed one, and what happened to those deployments after there was nobody left toonce nobody maintained them.

How an instance is put togethe…://github.com/llamaxyz/llama) – which matters more than it sounds, because; the hosted docs site went with the domain.

Who actually deployed one Because eEvery instance creation emits an event from a knownone factory address, theso adoption question has an exact answer rather than a marketing onecan be counted exactly. Reading the factory’s (http…ht, and the launch-day pair. That is a real but small install base, and it is worth readCreating an instance is not using honestly in both directions. Fourteen organisations is more than a failed product usually ge: several of the addresses above hold their deployment events, and several – Moonwell, Socket, Perennial, the DeFi Education Fund – are substantial. It is also far below the scale at which anothing after them, so a count at the factory measures decisions to try the framework, not organisations that governed with it. For how the framework becomes infrastructure other people build on, which is the outcomepares with others in the tooling market (/wiki/da…tooling/dao-tooling-overview) of that period was competing for. Two caveats belong with that count, and the page originally carried neither. The first is arithmetic: an earlier read of this page covered only Ethereum, Base and Arbitrum, and missed the fifteen instances on Optimism and seven on Polygon, which together are half the install base. The second matters more. Creating an instance is not using one: several of the addresses below hold their deployment events and nothing after them, so an adoption figure counted at the factory measures decisions to try the framework, not organisations that governed with it, see Aragon alternatives (/wiki/dao-governance/tooling/frameworks/aragon-alternatives) .

The framework outlived the company The most interesting entry in that census is the last one, and it does not say what this page first read into it. What happened to the deployments The Towns Mainnet instance (h…ama.xyz stopped serving a page. A; the factory on an immutable chain does not know or care whether its maintainer still exists, so the deployment went through. But taccepts deployments with no maintainer behind it. The instance has never been use…67097184c701899553b7) on Base, which this page previously cited as substantial adoption. What still executes belongs…grate its contracts into River, a fact the page was missing; Llama’s co-founder describe…/towns-protocol/llama-fork) . So tThe framework did not outlive its company: it was absorbpassed byto a customer, and what kept runningstill runs is that customer’s own govern… with an absent maintainer is still the failure class DAO tooling…uity) calls the dangerous one, and it cuts both ways. On the good side, nothing was taken away: e. Existing instances kept executi…o vendor to be reconstructed. What did stop is everything that was not on-chainEverything off-chain stopped. There is no interface, no ho… to. An organisation adopting athe framework in this state is choosing tonow becomes its own maintainer, whether or not it notices at the time.

How Caper approaches this CA caper's answer to the same exposure is structural rather than promissory, and it is visible in the contract split. Each caper's state tier – the treasury, the proposal record and the ballots – is deployed once and never migrated, so nothing abouttreasury, proposals and ballots live in a part of the contract that is published once and cannot be changed, so a caper's existingpast decisions do not depends on anyone maintainer being around. The logic tier that carries the pricing the software. The part that handles trading and execution plumbing can be replaced, on two paths the registry's own authorisation table keeps separate: an UPGRADE proposal (/wiki/governance/execution) passed by the $CAPER caper's own members, which reaches the registry through a method gated to the live logic component rather than to any key, and an operator backstop held by the protocol admin badge. Both land through the same propose-then-activate timelock, whose length is fixed when the registry is constructed and has no setter, so neither the badge nor a vote can shorten it; the Stokenet registry the site runs on was constructed with a delay of zero. What is absent is not the operator but the vendor: no maintainer's disappearance either freezes the tier or lets it change. The trade Llama's instances face, between an immutable contract and a maintained one, is put to a vote rather than settled by whoever is still employed. See what a caper is (/wiki/foundations/what-is-a-caper) for the shape of that spliteither by a vote of the members of the $CAPER caper or by the protocol operator. What is a caper (/wiki/foundations/what-is-a-caper) describes the design, and Execution (/wiki/governance/execution) covers what a passed proposal can do.

References AllThe repository figures on this page were read directly on 9 August 2026. The llamaxy…es-llama) of 1 November 2024.

How Caper approaches this Cap…d by the protocol admin badge.Both land through the same pro… for the shape of that split.

Category Onchain governance an…ty / Foundry) Built by Llama —– a DAO service provider whose …-chain 19 August 2023 Status 🔴 Wound down —Acquired, then unmaintained. River Protocol, now Towns, announced the acquisition of Llama on 1 November 2024 (https://blog.river.build/river-acquires-llama) ; llama.xyz resolves to nothing, and all 52 of the organisation'’s public repositories are arch…m, Base, Polygon, Blast, Zora —– one factory address (https:/…very chain Measured adoption 2143 instances creationsed across the threfive chains readable this run; ~14 distinct organisationswhose explorers answer – Optimism 15, Arbitrum 9, Base 7, Polygon 7, Ethereum 5 – read 19 September 2026 Audits Three Spearbit (http…/voting/openzeppelin-governor)

Llama was an onchain governanc…work for smart contracts, and —– separately —– the name of the firm that bui…. This page is not about that —– the shutdown, and the general… nobody left to maintain them.

Roles, not token weight Most …nce is a non-transferable NFT —– a policy —– and the policy encodes which …sury holds in other protocols —– the same problem Safe (/wiki…ted onto an existing treasury.

How an instance is put togethe…496EECE31083a8F493edd79AbEB) —– the identical address on all …//github.com/llamaxyz/llama) —– which matters more than it so…ocs site went with the domain.

Who actually deployed one Bec… a marketing one. Reading the factory'’s decod(https://eth.blockscout.com/address/0xFf5d4E226D9A3496EECE31083a8F493edd79AbEB) LlamaInstanceCreated logs on the threfive chains whose public explorers answered this run — Ethereum, Base and Arbitrum — gives 21 instance creations gives 43 instance creations (Optimism 15, Arbitrum 9, Base 7, Polygon 7, Ethereum 5, read 19 September 2026), and resolving each instance'’s on-chain name gives roughly …ourteen distinct organisations on the three chains named below. Several appear on more than …o", both dated 19 August 2023 —– the framework's first day liv…uct usually gets, and several —– Moonwell, Socket, Perennial, the DeFi Education Fund —– are substantial. It is also f…that period was competing for. Two caveats belong with that count, and the page originally carried neither. The first is arithmetic: an earlier read of this page covered only Ethereum, Base and Arbitrum, and missed the fifteen instances on Optimism and seven on Polygon, which together are half the install base. The second matters more. Creating an instance is not using one: several of the addresses below hold their deployment events and nothing after them, so an adoption figure counted at the factory measures decisions to try the framework, not organisations that governed with it.

The framework outlived the com…in that census is the last one, and it does not say what this page first read into it. The Towns Mainnet instance (https://eth.blockscout.com/address/0x343998c7056f75341c434e4db39a8be99a5ea4b4) was created on 3 March 2025: roughly, six and a half months after t…commit to any Llama repository, and around four months after …ama.xyz stopped serving a page at all. Somebody deployed a production governance instance from a framework whose authors had already left, and it worked, because a factory on an immutable chain does not kn. A factory on an immutable chain does not know or care whether its maintainer still exists, so the deployment went through. But the instance has never been used: its address carries its deployment events and no action in the eighteen months since, read on 19 September 2026. The same holds for Moonwell’s instance (https://base.blockscout.com/address/0x9ea8d23f0221a46152df67097184c701899553b7) on Base, which this page previously cited as substantial adoption. What still executes belongs to the acquirer. River Protocol, now Towns, announced on 1 November 2024 (https://blog.river.build/river-acquires-llama) that it had bought Llama and would integrate its contracts into River, a fact the page was missing; Llama’s co-founder described the same transaction (https://www.shreyashariharan.com/llama-is-acquired-by-river-protocol) from the other side. The busiest instance in the whole census is River’s own or care whether its maintainer still exists. This is the concrete case ofn Base (https://base.blockscout.com/address/0xa547373eb2b3c93adeb27ec72133fb7b92f70f7f) , which was still queuing and executing actions on 11 September 2026, and the only meaningful fork of the contracts is the acquirer’s (https://github.com/towns-protocol/llama-fork) . So the framework did not outlive its company: it was absorbed by a customer, and what kept running is that customer’s own governance. A live contract with an absent maintainer is still the failure class DAO toolin…uity) calls the dangerous one — a live contract with a dead maintainer —, and it cuts both ways. On the…or not it notices at the time.

How Caper approaches this Cap…plit. Each caper's state tier —– the treasury, the proposal record and the ballots —– is deployed once and never mi… for the shape of that split.

References All figures on thi…https://github.com/llamaxyz) —– 52 public repositories, every…//github.com/llamaxyz/llama) —– the framework repository, arc…496EECE31083a8F493edd79AbEB) —– the same address on all seven… from 30 November 2024 onward. The instance census was re-read on 19 September 2026 across five chains, counting LlamaInstanceCreated at the factory and then each instance’s own logs; the acquisition is dated from River’s announcement (https://blog.river.build/river-acquires-llama) of 1 November 2024.

How Caper approaches this Cap…e. Both land through the same timelock, and only the badge can set that timelock's length — governance can trigger the delay but never shorten itpropose-then-activate timelock, whose length is fixed when the registry is constructed and has no setter, so neither the badge nor a vote can shorten it; the Stokenet registry the site runs on was constructed with a delay of zero. What is absent is not the op… for the shape of that split.

How Caper approaches this Cap…ion plumbing can be replaced, but only byon two paths the registry's own authorisation table keeps separate: an UPGRADE proposal (/wiki/g…the $CAPER caper's own members and routed through the protocol's registry — there is no admin key that upgrades it and no vendor whose, which reaches the registry through a method gated to the live logic component rather than to any key, and an operator backstop held by the protocol admin badge. Both land through the same timelock, and only the badge can set that timelock's length — governance can trigger the delay but never shorten it. What is absent is not the operator but the vendor: no maintainer's disappearance either freezes ithe tier or lets it change. The trade … for the shape of that split.

Category Onchain governance and access-control framework (Solidity / Foundry) Built by Llama — a DAO service provider whose GitHub organisation (https://github.com/llamaxyz) dates to 12 November 2020 Framework launched Repository created 10 January 2023; first instances deployed on-chain 19 August 2023 Status 🔴 Wound down — llama.xyz resolves to nothing, and all 52 of the organisation's public repositories are archived (last push anywhere, 16 August 2024) Still deployed on Ethereum, Optimism, Arbitrum, Base, Polygon, Blast, Zora — one factory address (https://etherscan.io/address/0xFf5d4E226D9A3496EECE31083a8F493edd79AbEB) on every chain Measured adoption 21 instance creations across the three chains readable this run; ~14 distinct organisations Audits Three Spearbit (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Spearbit-Audit.pdf) reviews and one Code4rena (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Code4rena-Audit.md) ; the last is v1.1.0, January 2024 (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Spearbit-Audit-3.pdf) Licence MIT Related DAO tooling discontinuity (/wiki/dao-governance/concepts/analysis/dao-tooling-discontinuity) , Zodiac (/wiki/dao-governance/tooling/frameworks/zodiac) , Hats Protocol (/wiki/dao-governance/tooling/membership/hats-protocol) , OpenZeppelin Governor (/wiki/dao-governance/tooling/voting/openzeppelin-governor)

Llama was an onchain governance and access-control framework for smart contracts, and — separately — the name of the firm that built it. The firm spent 2021–2022 as one of the better-known DAO service providers, writing and shepherding proposals for Aave (/wiki/daos/lending/aave-dao) in particular; its repository list still reads as a record of that work, from the Aave treasury dashboard's ETL (https://github.com/llamaxyz/aave-etl) through an escrow between Friends With Benefits and Gamma Strategies (https://github.com/llamaxyz/fwb-liquidity-provisioning) to the Gitcoin–Radicle Public Goods Alliance (https://github.com/llamaxyz/gtc-rad-public-goods-alliance) . In January 2023 it turned the pattern it had been implementing by hand into a product: a framework for organisations that need to delegate narrow, revocable authority over a treasury rather than put every action to a token vote. The firm is gone. This page is not about that — the shutdown, and the general problem it belongs to, are covered on DAO tooling discontinuity (/wiki/dao-governance/concepts/analysis/dao-tooling-discontinuity) , where Llama is one of a measured cohort. What is documented here is what the framework did, who actually ran one, and what happened to those deployments after there was nobody left to maintain them.

Roles, not token weight Most onchain governance in this period was token-weighted: your influence is your balance, and every executable action passes through the same proposal queue. Llama started from the opposite premise. Membership of a Llama instance is a non-transferable NFT — a policy — and the policy encodes which roles the holder has. Permission is then a specific triple: a role, a target contract, and a function selector. A policyholder does not have "voting power"; they have the right to call this function on that contract, and nothing else. Action execution rules are modular. Rather than one quorum for everything, an instance attaches a strategy per permission, and the framework ships five: relative-quantity quorum, relative-holder quorum, relative-unique-holder quorum, absolute quorum, and absolute peer review. The same instance can therefore require a full role-wide supermajority to move the treasury while letting a two-of-five peer review approve a parameter change, without either rule leaking into the other. Funds are held in LlamaAccount contracts belonging to the instance, with a delegation-aware variant for governance tokens the treasury holds in other protocols — the same problem Safe (/wiki/dao-governance/tooling/treasury/safe) -based treasuries solve with modules. The design goal, stated in the repository, is programmatic control of funds: an approved action is a call the contracts make, not an instruction a multisig is trusted to carry out. The nearest live comparisons today are Hats Protocol (/wiki/dao-governance/tooling/membership/hats-protocol) for the role-as-token idea and Zodiac (/wiki/dao-governance/tooling/frameworks/zodiac) for modular execution rules bolted onto an existing treasury.

How an instance is put together Everything is deployed from a single LlamaFactory, which sits at 0xFf5d4E22…AbEB (https://etherscan.io/address/0xFf5d4E226D9A3496EECE31083a8F493edd79AbEB) — the identical address on all seven supported chains, the signature of a deterministic deployment. Each instance the factory creates is a LlamaCore (the action queue and execution engine), a LlamaPolicy (the role NFTs and the permission set), and an executor that performs approved calls. The strategy and account logic contracts are shared singletons the instances point at, so a new organisation deploys a small amount of code rather than a full framework. A read-only LlamaLens helper exists purely to make instances legible to interfaces, and two governance scripts batch the common multi-step administrative changes. The framework was reviewed four times before its last release: Spearbit in June 2023 (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Spearbit-Audit.pdf) , Code4rena (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Code4rena-Audit.md) , a second Spearbit in August 2023, and a third for v1.1.0 in January 2024 (https://github.com/llamaxyz/llama/blob/main/audits/Llama-Spearbit-Audit-3.pdf) . All four reports, and the framework's own documentation directory (https://github.com/llamaxyz/llama/tree/main/docs) , are still readable in the archived repository (https://github.com/llamaxyz/llama) — which matters more than it sounds, because the hosted docs site went with the domain.

Who actually deployed one Because every instance creation emits an event from a known factory address, the adoption question has an exact answer rather than a marketing one. Reading the factory's decoded logs on the three chains whose public explorers answered this run — Ethereum, Base and Arbitrum — gives 21 instance creations, and resolving each instance's on-chain name gives roughly fourteen distinct organisations. Several appear on more than one chain, which is one deployment decision rather than several. - Ethereum (5). Llama's own instance and a "Demo", both dated 19 August 2023 — the framework's first day live; the DeFi Education Fund (https://etherscan.io/address/0x585ebAC736B9d9A8Ee26C7A9fDa7a3987b4ff517) , whose Uniswap proposal work has its own repository (https://github.com/llamaxyz/def-uniswap-proposal) in the same organisation; MYSO; Daylight; and Towns Mainnet (https://etherscan.io/address/0x343998c7056f75341C434E4db39A8be99A5EA4B4) . - Base (7). Moonwell (https://basescan.org/address/0x9Ea8D23f0221A46152dF67097184C701899553B7) , Socket, River, Drakula, Daylight, plus the launch-day Llama and Demo instances again. - Arbitrum (9). Perennial (https://arbiscan.io/address/0x635DE406f437911BCefFE6973CeD931A4ceb6cce) , RabbitHole, a LI.FI test instance, three separate Socket deployments, Daylight, and the launch-day pair. That is a real but small install base, and it is worth reading honestly in both directions. Fourteen organisations is more than a failed product usually gets, and several — Moonwell, Socket, Perennial, the DeFi Education Fund — are substantial. It is also far below the scale at which a framework becomes infrastructure other people build on, which is the outcome the tooling market (/wiki/dao-governance/tooling/dao-tooling-overview) of that period was competing for.

The framework outlived the company The most interesting entry in that census is the last one. The Towns Mainnet instance was created on 3 March 2025: roughly six and a half months after the final commit to any Llama repository, and around four months after llama.xyz stopped serving a page at all. Somebody deployed a production governance instance from a framework whose authors had already left, and it worked, because a factory on an immutable chain does not know or care whether its maintainer still exists. This is the concrete case of the failure class DAO tooling discontinuity (/wiki/dao-governance/concepts/analysis/dao-tooling-discontinuity) calls the dangerous one — a live contract with a dead maintainer — and it cuts both ways. On the good side, nothing was taken away: existing instances kept executing, the code is MIT-licensed and readable, the audit reports survived in the repository, and the on-chain record needs no vendor to be reconstructed. What did stop is everything that was not on-chain. There is no interface, no hosted documentation, no upgrade path, and no one to write a patch: the security review a prospective adopter would rely on is frozen at v1.1.0 in January 2024, and any vulnerability found after that date has no maintainer to report it to. An organisation adopting a framework in this state is choosing to become its own maintainer, whether or not it notices at the time.

How Caper approaches this Caper's answer to the same exposure is structural rather than promissory, and it is visible in the contract split. Each caper's state tier — the treasury, the proposal record and the ballots — is deployed once and never migrated, so nothing about a caper's existing decisions depends on a maintainer being around. The logic tier that carries the pricing and execution plumbing can be replaced, but only by an UPGRADE proposal (/wiki/governance/execution) passed by the $CAPER caper's own members and routed through the protocol's registry — there is no admin key that upgrades it and no vendor whose disappearance either freezes it or lets it change. The trade Llama's instances face, between an immutable contract and a maintained one, is put to a vote rather than settled by whoever is still employed. See what a caper is (/wiki/foundations/what-is-a-caper) for the shape of that split.

References All figures on this page were read directly on 9 August 2026. The llamaxyz GitHub organisation (https://github.com/llamaxyz) — 52 public repositories, every one flagged archived, latest push anywhere 16 August 2024. llamaxyz/llama (https://github.com/llamaxyz/llama) — the framework repository, archived, MIT, created 10 January 2023. LlamaFactory on Ethereum (https://etherscan.io/address/0xFf5d4E226D9A3496EECE31083a8F493edd79AbEB) — the same address on all seven deployment chains; instance counts and dates come from its decoded LlamaInstanceCreated logs on Ethereum, Base and Arbitrum, and each instance's name from an eth_call to its LlamaCore. The domain's disappearance is dated from the Internet Archive's index for llama.xyz, which records HTTP 200 responses through 10 October 2024 and 404s from 30 November 2024 onward.