please note that a few fields have ‘…to fill in’ mentions, these will be edited in within the next few days.
SECTION 1: APPLICANT INFORMATION
Provide personal or organizational details, including applicant name, contact information, and any associated organization. This information ensures proper identification and communication throughout the grant process.
Applicant Name:
Marc Tillement
Project Name:
Pyth Network
Project Description:
The Pyth Network is a first-party financial oracle network designed to provide low-latency market data to multiple blockchains securely and transparently.
Pyth currently supports 400+ real-time price feeds across crypto, equities, ETF’s, FX pairs, and commodities, and has facilitated more than $100B in total trading volume across over 50 blockchain ecosystems.
Team Members and Roles:
Outside of external contribution, there are 2 teams contributing to the Pyth Network as of today: Pyth Data Association and Douro Labs.
You can find the Pyth Data Association team here.
You can find the Douro Labs team here.
Project Links:
All relevant links can be found here: pythnetwork | Twitter | Linktree
Do you acknowledge that your team will be subject to a KYC requirement?:
Yes
SECTION 2a: Team and Product Information
Provide details on your team’s past and current experience. Any details relating to past projects, recent achievements and any past experience utilizing incentives. Additionally, please provide further details on the state of your product, audience segments, and how you expect incentives to impact the product’s long-term growth and sustainability.
Team experience (Any relevant experience that may be useful in evaluating ability to ship, or execution with grant incentives. Please provide references knowledgeable about past work, where relevant. If you wish to do so privately, indicate that. [Optional, but recommended]):
The success of the project itself speaks to the team’s capabilities. The Pyth Data Association (PDA) and Douro Labs (DL) teams have been developing Pyth Network for almost 3 years now. Pyth launched its cross-chain oracle in December 2022 and over the past year has grown to be the second-most integrated oracle in the world. The product itself powers numerous protocols across different verticals, including perps, borrow/lending, and stablecoins.
- The PDA team has led the ecosystem growth to more than 90 data publishers and more than 170 applications powered by Pyth (including the close collaboration with Vela, HMX and Synthetix). The team has also enabled the growth of the Pyth community to 70k members and more than 160k wallets of $PYTH stakers. Individuals responsible for this success include : Marc, Pierre, Ed, Mario, Ani, Ali, Guillermo, and other contributors from Douro Labs or the Pyth Data Association
- The Core Engineering team at Douro Labs, led by Jayant, has led the build out of the Pyth protocol from inception, the v1 delivery using aggregation on-chain on Solana and expanding it to an appchain (Pythnet) with the ability of low-latency delivery of data to multiple chains (53 to date)
- Pyth Data Association and Douro Labs have delivered data feeds that are unmatched in performance (updates every 400ms at subsecond latency) and user protection through working closely with application to implement commit-and-reveal schemes that protect against front running. Below are examples of how application on Arbitrum have used Pyth:
On designing and implementing effective incentives, The Pyth Data Association recently performed the Pyth Network Retrospective Airdrop, which allocated 2.5% of its token supply to on-chain users of #PoweredByPyth applications and 1% directly to protocols being #PoweredByPyth. The airdrop primarily recognized usage of Pyth price feeds by applications and extended the rewards to users of such applications.
Airdrop Phase 1 dropped PYTH tokens to DeFi users across 27 chains. Then, this airdrop was the largest cross-chain airdrop ever performed — over 52K wallets have claimed their PYTH tokens. Airdrop Phase 2 rewarded over 160 DeFi protocols using the Pyth Price Feeds.
The Pyth Network Retrospective Airdrop was highly successful in galvanizing and growing the Pyth community (400K PYTH token holders today). The Airdrop was also covered in various media outlets (Coindesk, or The Block among others), and close to 200K wallets are staking PYTH in governance.
What novelty or innovation does your product bring to Arbitrum?
The Pyth Network oracle is a permissionless pull oracle offering about 450 price feeds on Arbitrum. The price feeds range from crypto assets to FX, metals, equities, ETF and US rates.
- Permissionlessness: no need to contact the team, just read the documentation and become powered by Pyth
- Pull oracle: do not comprise on price accuracy, if you need to know the current price on-chain, trigger a price update
- Extensive price feeds catalogue: whether you’re building a crypto-focused platform or RWA one, you have access to 450 price feeds today — and the catalogue keeps growing!
The ability of multiple applications to trigger price updates on-demand enables applications to avoid front and back-running attacks that emanate from applications using push-model oracles, where the price on-chain is already known. The pull-model also allows applications to implement commit-and-reveal schemes that reduce the number of transactions on-chain (including price updates) to those related to a trade committed to by the user.
The ability to subsidise the costs related to such price updates will enable multiple applications to trigger a higher number of updates independently and allow their users to use prices with the highest levels of freshness and reduce the risk of front-running
Is your project composable with other projects on Arbitrum? If so, please explain:
The Pyth Price Feeds are by nature composable as it can power DeFi applications. Additionally, the 400+ Pyth prices are permissionlessly available to anyone — no need for API keys or permissions.
Various leading Arbitrum applications are already using Pyth: HMX, Vela, Perennial, Synonym, KTX Finance, WINR, Cega, Deri, Rage Trade, Pingu Exchange, Pike Finance, Tigris
LTIPP applicants like Synthetix or KTX and others would further grow the Pyth oracle footprint on Arbitrum.
Do you have any comparable protocols within the Arbitrum ecosystem or other blockchains?
Chainlink, Redstone, Stork, and others according to Defillama.
As of today, Pyth is the second most used oracle on Arbitrum after Chainlink.
How do you measure and think about retention internally? (metrics, target KPIs)
Being a permissionless pull/on-demand oracle powering applications on multiple chains, one of the key metric for growth and retention is the number of applications powered by Pyth, the number of developers and the associated size of activities performed by those applications. As can be verified by the chart below, the number of applications and associated developers using Pyth services has increased constantly, exceeding any churn from applications ceasing operations.
To further demonstrate usage of Pyth data, the chart below shows the number of (daily) price updates by protocols that need this information to operate — Every Pyth price update on-chain is deemed related to the execution of a transaction on the application where the user triggered the update and the application committing the data on-chain (for a gas fee of $2 as of end of February).
Relevant usage metrics - Please refer to the OBL relevant metrics chart 23. For your category (DEX, lending, gaming, etc) please provide a list of all respective metrics as well as all metrics in the general section:
- Daily Active Users: As of today, about 230 distinct smart contracts have integrated and interacted permissionlessly with the Pyth oracle (triggering or reading price updates). Until now, over 7,500 unique users have interacted with an application powered by Pyth.
- Daily User Growth: Deployed in Q3 2022, the Pyth smart contract saw 169 smart contracts integrating by end of 2023. In the past 2.5 months, an additional 57 contracts started using the Pyth oracle. Until the end of 2023, around 5,000 wallets had interacted with a smart contract powered by Pyth. Since the start of 2024, the number grew 50% ; from 5K to 7.5K
- Daily Transaction Count: Since deployment, a total of 475,000 price updates were permissionlessly triggered by any of these 230 smart contracts. There were a total of 207,000 price updates done in 2023 (about 567 / day), and since the beginning of 2024 (2.5 months), the number of price updates triggered is 270,000 (or 3,5000 / day), a 6x increase compared to 2023.
- Daily Protocol Fee: As of today, the Pyth oracle collects 1 wei for every price feed updated on-chain. Therefore the Pyth protocol fee is equal to the daily transaction count mentioned above (1 wei x number of updates).
- Daily Transaction Fee: Over the last 15 months, over 1850 $ETH have been spent by applications powered by Pyth to update price feeds and effectively use that price (trade, liquidations etc). Since 2024, 650 ETH have been spent (8 ETH / day or ) to do the requested price updates (270,000).
- Incentivized User List & Gini: The Pyth Activity by apps Dune query shows all contracts that updated a price feed in 2024, and display the number of updates triggered as well as the gas cost incurred to do so (from 2024 also). Once LTIPP starts, dates will be updated to match the grant period and this data will be used as reference for the weekly refunding of the cost incurred to update the Pyth price feeds.
- Total Value Secured: As of today, the total value secured by the Pyth oracle on Arbitrum is around $175M.
- Total Trading Volume enabled: Pyth has enabled over $40B in trading volume on various Arbitrum decentralized perpetual exchanges (HMX, Vela, WINR, CAP, KTX, Level Finance, SynFutures, Pingu Exchange, Perennial, Gambit, Deri V4, Buffer, Cryptex Finance and Tigris Trade).
Dune Dashboard: https://dune.com/cctdaniel/pyth-arbitrum-ltipp
Do you agree to remove team-controlled wallets from all milestone metrics AND exclude team-controlled wallets from any incentives included in your plan:
Yes. The usage of this LTIPP will be activity based and aims to offset the cost incurred by applications.
Did you utilize a grants consultant or other third party not named as a grantee to draft this proposal? If so, please disclose the details of that arrangement here, including conflicts of interest (Note: this does NOT disqualify an applicant):
No, this was drafted by contributors from the Pyth Data Association with the help from Douro Labs.
SECTION 2b: PROTOCOL DETAILS
Provide details about the Arbitrum protocol requirements relevant to the grant. This information ensures that the applicant is aligned with the technical specifications and commitments of the grant.
Is the protocol native to Arbitrum?:
No.
On what other networks is the protocol deployed?:
The Pyth oracle contract is deployed on over 50 chains (mainnet), ranging from EVM, to Cosmwasm, Move chains and Solana or NEAR.
You can find all EVM contracts here, Move ones, Cosmwasm here, or Solana and NEAR in our documentation.
What date did you deploy on Arbitrum mainnet?:
The Pyth oracle contract has been deployed on Arbitrum on Sep-09-2022 03:56:19 PM +UTC with this tx.
Do you have a native token?:
Yes the Pyth Network has a governance token: $PYTH.
For tokenomics please refer to the documentation and for the initial token launch, read on here.
To find all PYTH token addresses (including on Arbitrum), refer to the docs.
Past Incentivization:
The Pyth Network has had to date, two major ‘incentivization’ events:
- Pyth Retrospective Airdrop: Phase 1 - DeFi Participants
- Pyth Retrospective Airdrop: Phase 2 - Decentralized Apps
The phase 1 allocated 2.5% of PYTH total supply to end-users of applications powered by Pyth. This involved 90K wallets from 27 different chains. Over 2/3 of the airdrop has been claimed almost 4 months after the launch.
The phase 2 allocated and distributed 1% of the PYTH token supply directly to applications/DAOs integrated with the Pyth oracle.
These 2 phases have lead close to 200K unique wallets to stake (1.2B) PYTH tokens in governance. After Ethereum, Solana, and Celestia, the Pyth Network is the 4th biggest network in distinct stakers.
The Pyth Network has never received grants incentivizing Pyth usage from any ecosystem it is available on.
Current Incentivization:
As of today (past Phase 1 and 2 of the Pyth Retrospective Airdrop), only Pyth Data Publishers are rewarded for publishing data to the network. Rewards are allocated to each publishers based on this ranking which depend on the number on which a publisher is active (with enough quality data).
No grants were received from the Arbitrum foundation nor any participant in the Arbitrum ecosystem.
Protocol Performance:
According to DefiLlama, over 30 applications are powered by Pyth on Arbitrum.
Additionally, and as mentioned before, being a permissionless pull/on-demand oracle, demonstration of usage are the number of (daily) price updates by protocols that need this information to operate properly.
Specifically for Arbitrum, the oracle is now delivering about 4,000 to 6,0000 price updates daily to apps powered by Pyth (according to the Dune Dashboard).
In regards to trading volume enabled, the Pyth Price Feeds already secured over $40B in trading volume. Looking further than Arbitrum, over $120B trading volume has been enabled thanks to the Pyth Price feeds.
Protocol Roadmap:
The Pyth Network is operated by the Pyth DAO, and so future endeavors will have to be decided/approved by the PYTH stakers and the Pyth DAO Councils.
Nonetheless, contributors such as the Pyth Data Association and Douro Labs have focused on 2 new endeavors:
- Pyth Entropy — Pyth Entropy allows developers to quickly and easily generate secure random numbers on the blockchain for applications such as NFT mints, gaming, and more. Pyth Entropy is currently available on several EVM networks.
- Pyth Express Relay — enables an efficient and secure liquidation process for DeFi applications that traditionally have had to share most of liquidation revenues with block miners. This protocol is already deployed on testnet and both applications and searchers have started testing it.
Audit History & Security Vendors:
The Pyth oracle has undergone many and thorough audits from reputable companies such as OtterSec, Trail of Bits and Asymmetric Research.
You can find all Pyth related audits here, they involve either the oracle itself as well as the Pyth DAO and governance aspect of the network. Within the past 2 years, 14 audits have been done.
A bug bounty program also exists, rewarding up to $500K for critical finds — you can find all the details (in-scope as well as process) here.
Security Incidents:
Neither the oracle nor governance smart contracts were ever exploited.
Regarding the Pyth Price Feeds themselves (and an inaccurate price to be produced), we’d need to go back to September 2021 (2 months after the V1 launch).
You can find the full RCA on our blog. Issue was since fixed on the code side and the whole reliability and benchmarks used by the oracle have since been massively upgraded. Please refer to the blog; Reliability at Pyth, for discovering Pyth security practices for price feeds.
SECTION 3: GRANT INFORMATION
Detail the requested grant size, provide an overview of the budget breakdown, specify the funding and contract addresses, and describe any matching funds if relevant.
Requested Grant Size:
1,000,000 ARB (revised post Dencun)
Justification for the size of the grant:
The Grant will be used to improve the access to price feeds for Arbitrum-based DeFi apps at no cost — this will enable DeFi applications to a. increase the number of assets supported on their protocol, b. alleviate the cost burden from users or dApps, c. further optimize applications UX (less fees charged to users, more granular price updates).
A Pyth price update cost about $1 (we currently see a moving average of $0.8 post Dencun, down from $5 per-dencun, but expect it to slightly increase) to apps or end-users, so with the goal to reach 25,000 price updates per day, this would translate to a spending of $25,000 daily in gas fees (or 2,100,000$ for the 12 weeks duration of the LTIPP).
$2,100,000 would translate to 1,240,000 ARB at today’s prices (lower than the requested amount). With the Dencun upgrade now live and the transaction costs remaining under $1 for now (only 3 days of history), we believe the gap is reasonably within the measurement error.
Grant Matching:
The PDA has a strong history of incentivizing both Data Publishers and DeFi Users. The Pyth Retrospective Airdrop (Phase 1 and 2) has rewarded DeFi applications and users using Pyth on Arbitrum. Within the 2 phases, over 25M PYTH were distributed to Arbitrum users and applications.
Specifically part of this LTIPP, the Pyth Data Association has not planned to match any grant amount received ; one of the reason being that apps like Vela or HMX have recently received significant Pyth retrospective airdrop.
Grant Breakdown:
1,000,000 for gas fees update rebates — this will be activity based and disbursed on a weekly basis
Funding Address:
0x4133F561B747F6cd2ECd4f03515A632F5e4D8C8D
Funding Address Characteristics:
2/3 multisig, signers are contributors to the Pyth Data Association
- Marc Tillement, Director
- Giulio Alessio, Head of Operations
- Edward Zuo, Head of Growth
Treasury Address:
The Pyth DAO funds can be found here: kiXuGDcVfXpFqKw2oKMnwbi1nJaGayqRBXucgQGe9GK — this is a Solana SPL address
Additionally, any funds held in any contracts owned by the DAO are its property. For instance, on Arbitrum, the Pyth DAO also owns all the funds in this address: 0xff1a0f4744e8582df1ae09d5611b887b6a12925c
Contract Address:
0xff1a0f4744e8582DF1aE09D5611b887B6a12925C
SECTION 4: GRANT OBJECTIVES, EXECUTION AND MILESTONES
Clearly outline the primary objectives of the program and the Key Performance Indicators (KPIs), execution strategy, and milestones used to measure success. This helps reviewers understand what the program aims to achieve and how progress will be assessed.
Objectives:
The grant aims to use ARB to implement a rebate program to applications for gas fees related to Pyth price updates on Arbitrum.
Execution Strategy:
Allocation of ARB incentives within a total of 12-weeks for the improvement of performance and access to Pyth price feeds:
1,000,000 ARB to implement a rebate program to DeFi applications on Arbitrum for gas fees related to price updates. the distribution will incentivise both existing and new consumers of the Pyth data and gradually increase as the number of updates committed on-chain increases. the program will last for 12 weeks with a 4-week ramp-up to engage with existing and new applications.
What mechanisms within the incentive design will you implement to incentivize “stickiness” whether it be users, liquidity or some other targeted metric?
To enhance stickiness of activity and ensure sustained engagement with Pyth on Arbitrum, the incentive design includes multiple strategies:
- Gas rebates incentives combined with application co-marketing campaigns will be used to seed critical mass adoption of Arbitrum-based applications powered by Pyth Price Feeds, and on-going engagement over time;
- Incentives to Pyth community users to use Arbitrum-based applications enables pollination of new usage using a community that is using DeFi applications on other chains and expands the pool of dedicated users on Arbitrum;
- Emphasis on Continuous Product Innovation through the increase of performance, reduction of latency and deployment of innovative protocols that enable applications to improve UX and compete;
- Encourage involvement in Governance through gradual delivery of rewards and the fostering of deeper connections with the Arbitrum ecosystem. Pyth has already one of the highest staking ratios in all of Web3, and we expect to onboard many new applications and users to use Price Feeds through ARB rewards; and
- While the initial incentives aim to accelerate the bootstrapping phase, Pyth has built in a sustainable business model that removes the ongoing reliance on such incentives and that allows for Applications and Data Publishers to be incentivised in participation over the long-term.
Specify the KPIs that will be used to measure success in achieving the grant objectives and designate a source of truth for governance to use to verify accuracy.
The various KPIs for the LTIPP are:
- Daily Active Users (smart contracts): As of today, about 230 distinct smart contracts have integrated and interacted permissionlessly with the Pyth oracle (triggering or reading price updates). With the incentivizes in places and the growing Arbitrum developers ecosystem, goal is to reach 750 by the end of the LTIPP.
- Daily Active Users (end users): Until now, over 7,750 unique users have interacted with an application powered by Pyth. With the LTIPP, there will be more powered by Pyth apps and those will be more competitive in offering the best UX to users ; thus, we’re aiming for over 40,000 distinct users at the end of LTIPP.
- Daily Transaction Count: There were a total of 207,000 price updates done in 2023 (about 567 / day), and since the beginning of 2024 (2.5 months), the number of price updates triggered is 270,000 (or 3,500 / day), a 6x increase compared to 2023. Through this LTIPP, the goal is to have 25,000 updates / day, or in other words, an increased activity by 2x from the last know peak (13k updates on March 16th).
- Total Value Secured: By the end of the LTIPP, goal is to have $500M of TVL secured by the Pyth Price Feeds. TVL metrics are predominantly driven by money-market protocols (like Aave, Radiant or Silo) while Pyth footprint on Arbitrum til date is massively geared towards derivatives (that require less TVL). Thus, goal is that by the end of LTIPP the Pyth Price Feeds would secure a leading money-market on Arbitrum.
- Total Trading Volume enabled: With LTIPP (and so offsetting the gas costs to dApps and to end users), we’re expecting powered by Pyth applications to the main and best venues for trading derivatives contract. With $40B enabled over the last 18 months, the goal is to enable $50B by the end of 2024 (a 300% — from today’s average of +$2B monthly volume to over $6B monthly)
The 1st 3 KPIs can be checked directly from this Dune Dashboard. Specifically, DAU here, DUG here, and DTC here. For the last 2 KPIs: TVS and Trading Volume enabled, we’ll provide consolidated metrics ad hoc with the relevant proof (for instance TVL secured by a powered by Pyth application from DefiLlama or stats/analytics page of a perpetual protocol using Pyth). The latter 2 KPIs will be provided via a Dune Dashboard but will require to be built (TO DO).
Grant Timeline and Milestones:
Month 1-2 (April 20th-May 29th): Deployment and Ramp-Up Period
- Advertise and establish the ability of existing and new Arbitrum-based applications to claim ARB to claim gas rebates
- KPIs to watch: number of applications using Pyth Price Feeds, number of price feed updates per day
- Advertise and establish the ability of existing and new users of Arbitrum-based applications to claim ARB weighted by activity on Arbitrum
- KPIs to watch: number of new users using Arbitrum-based applications, number of price feed updates per day
Month 2-3 (May 20th - July 19th): Acceleration and Volume
- Push for 20 new price feeds used by Arbitrum-based applications
- KPI to watch: number of price feeds available on Arbitrum-based applications
- Rebate 100% of gas fees to all Arbitrum-based applications
- KPI to watch: amount of gas used in price feed updates rebated to applications
- 10,000 new users from the growing Pyth Community are active new users on Arbitrum
- KPI to watch: the increase of Pyth members using Arbitrum, volume of transaction that Pyth community members are contributing to Arbitrum
How will receiving a grant enable you to foster growth or innovation within the Arbitrum ecosystem?
Receiving this grant will significantly impact the Arbitrum ecosystem by driving growth and fostering innovation in several ways:
- New applications and new price feeds on Arbitrum: By rebating gas fees, the grant will help incentivise applications operating on other-chains to adopt Arbitrum as their new/additional deployment
- Increased Application Activity: The incentive program is expected to boost DeFi activity on Arbitrum, further increasing the transaction activity on Arbitrum and establishing Arbitrum as a leading L2 ecosystem for DeFi
- User Adoption: With targeted incentives for new users from existing Pyth community and beyond, the program aims to significantly increase the number of users interacting with Arbitrum-based applications
- Experimentation and Innovation: the deployment of new price feeds on Arbitrum will benefit the Arbitrum ecosystem and the composability of protocols in the ecosystem
- Ecosystem Synergies: The growth of the Pyth ecosystem will impact applications driving overall growth. The increased catalogue of price feeds, size and diversity of the Pyth community will foster higher demand on Arbitrum
- Pyth Support for Arbitrum: Pyth, a pillar protocol in DeFi, encourages its community members to try out Arbitrum-based applications. Many applications have also shown interest in adopting Arbitrum, so this grant signals a strategic expansion that benefits both ecosystems.
Do you accept the funding of your grant streamed linearly for the duration of your grant proposal, and that the multisig holds the power to halt your stream?
Yes
SECTION 5: Data and Reporting
OpenBlock Labs has developed a comprehensive data and reporting checklist for tracking essential metrics across participating protocols. Teams must adhere to the specifications outlined in the provided link here: Onboarding Checklist from OBL 42. Along with this list, please answer the following:
Is your team prepared to comply with OBL’s data requirements for the entire life of the program and three months following and then handoff to the Arbitrum DAO? Are there any special requests/considerations that should be considered?
Yes. And no specific considerations to account for today.
Does your team agree to provide bi-weekly program updates on the Arbitrum Forum thread that reference your OBL dashboard?
Yes. The Pyth oracle being a permissionless pull model (meaning all price updates are shared/public) means that anyone can access the usage/metrics data directly on-chain. All metrics/goals stated in this LTIPP are (or will be) publicly verifiable, and it will be done so via Dune dashboards (or equivalent).
*First Offense: In the event that a project does not provide a bi-weekly update, they will be reminded by an involved party (council, advisor, or program manager). Upon this reminder, the project is given 72 hours to complete the requirement or their funding will be halted.
Understood.
Second Offense: Discussion with an involved party (advisor, pm, council member) that will lead to understanding if funds should keep flowing or not.
Understood.
Third Offense: Funding is halted permanently
Understood.
Does your team agree to provide a final closeout report not later than two weeks from the ending date of your program? This report should include summaries of work completed, final cost structure, whether any funds were returned, and any lessons the grantee feels came out of this grant. Where applicable, be sure to include final estimates of acquisition costs of any users, developers, or assets onboarded to Arbitrum chains. (NOTE: No future grants from this program can be given until a closeout report is provided.)
Yes.
Does your team acknowledge that failure to comply with any of the above requests can result in the halting of the program’s funding stream?:
Yes
