When the year started, the market was fatigued from the damage of the FTX collapse, as well as a barrage of macroeconomic, geopolitical, and regulatory events. The ongoing Russia-Ukraine War continued to have massive implications that contributed to global uncertainty and financial sector turmoil. Lingering effects from the COVID outbreak were still contributing to instability in the financial markets and across global supply chains. To add more fuel to the fire, the banking crisis that arose from Silicon Valley Bank triggered concerns about financial contagion within the banking, tech, and crypto sectors. To top it all off, the Fed implemented fiscal policy that brought us to the highest rates the market had seen in 22 years. 

While the state of global markets in 2023 appeared bleak, crypto experienced its own tribulations, ranging from a USDC de-peg (thank you SVB) to threats of regulatory action. 

Although sentiment across the board screamed “bear market,” the yearly charts whispered a different story. BTC and ETH are up 157% and 100%, respectively, and one could argue that crypto has actually been one of the best performing asset classes this year. Most surprisingly, the shift of VC attention to L2s left many unprepared for the revival of the monolithic Alt L1 thesis. Solana stole the stage in the last quarter of the year, rallying 700% YoY and setting new all-time highs in DEX volume. This sentiment lifted the tide for other Alt L1s, like Avalanche and Sei.

In conjunction with exciting technological developments across every major ecosystem, Blackrock and other institutional players entered the scene with applications for spot BTC and ETH ETFs. Given crypto’s sub $2T market capitalization, this potential influx of institutional flows ignited a flame that sent crypto asset prices flying higher into the end of the year. On top of this, the crypto user experience significantly improved through innovations in account abstraction, L2s, and cross-chain stablecoins. Large scale airdrops, healthy levels of venture funding, and the return of builder-entrepreneur excitement aided crypto in reaching pre-FTX levels. 

Recent price action coincided with substantial momentum across nearly every blockchain, warranting plenty of analysis for Blockworks Research’s 2023 Year End Review, Part 1.

So let’s dive in. 

Bitcoin Update

This year was one of the more eventful years in Bitcoin’s history between price volatility, hype over a spot ETF approval, the renewed interest in BTC from notable investors in traditional finance, the inception of Ordinals, and the network’s improving fundamentals as the industry looks to put the brutal bear market in the rearview mirror. As is typical in bear markets, BTC has enjoyed stronger performance relative to many other crypto assets, with BTC dominance rallying from ~42% to start the year to a high of ~54% at the end of October. We also saw an all-time low in the number of active BTC in the past year, which exemplifies the high conviction nature of BTC investors. 

This year also marks the last full calendar year of a 6.25 BTC block reward, with the halving slated for April 2024. The Bitcoin Network’s hash rate has continually notched all-time highs throughout the year, perhaps as a result of delayed ASIC shipments from 2022 finally being delivered, the continual improvements in ASIC computing power, or miners trying to capture the higher BTC-denominated returns ahead of the halving. Nevertheless, the network remains as secure as ever with over 460M TH/s of compute keeping the chain honest. The halving is likely to act as a major headwind for miners, who are still in a similar position as they were back in the depths of the bear market when BTC was trading at this cycle’s lows. This is due to the increased hash rate, and thus competition, to mine a block despite the higher BTC spot price. The following chart highlights the estimated revenue per day for a miner with 1M TH/s of operational computing power over 2023. As you can see, the revenue of a miner that hasn’t continually invested in more hardware throughout 2023 has barely increased its top line. 

A large driver for the spike in miner revenue during May, as well as more recently in November, can be attributed to Bitcoin Ordinals. The “Ordinal Theory” was popularized by Casey Rodarmor earlier this year, which involves inscribing individual satoshi’s with arbitrary data that inherit Bitcoin’s immutable, secure, and decentralized properties. Over $80M has been spent on fees to inscribe ~40M ordinals throughout 2023, with all fees flowing to miners. Ordinals can be data heavy, thus making blockspace more scarce and driving up transaction fees for users executing standard transfers. So while $80M isn’t going to move the needle on its own, the effect it has on overall transaction fee revenue for miners is rather significant when Ordinal inscriptions rise past a certain threshold. 

Ethereum

Ethereum as a network and ETH as an asset had a pretty interesting year with many ups and downs. While ETH has rallied 100% YTD, relative to BTC, it has underperformed by 57%. This is atypical, as ETH’s beta has historically been positively larger, which partly derives from the simple fact that the asset is smaller in comparison to BTC. Given this year’s major events, such as a flight to alternative stores of value due to the banking crisis and the anticipation for an imminent BTC ETF, ETH did not see many of the same tailwinds.

Although it was overshadowed by BTC at the end of 2023, ETH the asset and its narrative of “Ultrasound Money” was strong for a large majority of the year, as a net of ~350K ETH was burned from the supply over the course of the year. Despite the summer and early fall having a lull in activity, leading to some net inflation, supply continued to make local lows in December. 

Outside of ETH the asset, there was a lot to pay attention to throughout the year with respect to Ethereum Mainnet. The first major development in 2023 came with the Shapella (Shanghai and Capella) upgrade, which introduced the ability for ETH stakers to withdraw their assets from the Beacon Chain. Prior to this point, users had to lock their ETH for some indefinite period of time. Now, users can withdraw their stake whenever they please but have to wait until they are progressed through a queue with a maximum churn limit.

While many thought the introduction of withdrawals would lead to a mass selling event as the once-locked supply became available, the opposite occurred; those who had locked their ETH wanted long-term exposure beyond any date of withdrawals, and the risk of staking was now heavily reduced. As a result, the exit queue quickly vanished to zero as the entry queue climbed higher over the summer months. For more details on how the queue works, check out our report on the Merge. 

This growth in staked ETH can be largely attributed to the growth of Liquid Staking Tokens (LSTs)—liquid assets that represent the value of a staked ETH that pay out the staking yield either through rebasing the supply (e.g., stETH) or by having the ability to exchange the asset for more ETH from the protocol (e.g., wstETH, rETH, frxETH). LSTs enable higher capital efficiency by allowing the staked asset and yield to be utilized elsewhere and take the complexity of running a node out of users’ hands. With withdrawals enabled, arbitraging LSTs has become possible, leading to better price parity. The reduction in risk of utilizing LSTs alongside their capital efficiency and a low network stake rate was a recipe for exponential growth. 

The rise in stake, particularly from LSTs, introduced certain issues within the Ethereum community. As Lido, the dominant LST provider, continued to gain market share and started closing in on 33% of the total staked ETH, many raised concerns around centralization. If one entity can control this much stake, it means the ability to halt the finality of the chain. With additional collusion to 50% of the stake, it could lead to the ability to decide the correct fork choice and explicitly censor transactions. 

Those who defend Lido claim that it is not one entity but rather a group of many separate and independent node operators. This decentralization will only continue with the introduction of their staking router, which will include elements of a permissionless validator set and distributed validator technology (DVT). In addition, Lido’s governance gives veto power to stETH holders, allowing them to veto any decision made by the DAO. A Pareto distribution seems likely in the market of LSTs, as more trust will be given to those with greater liquidity, and the Lindy effect plays an important role here. What’s most important is that the underlying design is robust. 

While some in the community have called for the “self-limiting” of Lido, the protocol has no economic incentive to do so, and token holders have already expressed their view that they do not want to self-limit. Others are attempting to “vampire attack” the project by using incentives to convince users to deposit stETH into their protocol, which is then converted to a new LST on the back end. Diva is a public example of this, and we can expect others to try something similar. Either way, as Lido continues to maintain dominance, we will see initiatives from the community to siphon market share to even the playing field and reduce concerns around censorship. Regardless of which method you believe in, the ability for the community to openly discuss these ideas and raise concerns is a benefit of the systems we are creating.

As we head into 2024, there are a few main areas of focus for Ethereum developers, researchers, and the broader community. The first priority is the Dencun (Cancun and Denub) upgrade, which will include EIP-4844, also known as Proto-Danksharding. This will introduce the idea of state “blobs,” an optional 125KB sidecar on Ethereum transactions that will have its own separate fee market. This is where rollups that use Ethereum for DA or settlement will post their transaction data, as it will likely be orders of magnitude cheaper than the current method of posting calldata. 

While current estimates of savings seem to be between 80-90% for rollups to post their data, we cannot be sure of what these numbers look like until the upgrade goes live as cost savings are a result of supply and demand. These savings will be directly applied to users transacting on these L2s, bringing a level of scalability and enhanced user experience across the entire landscape.

Other developments to watch out for include EigenLayer, which looks to launch on Mainnet in the first half of 2024. With that will come the rise of liquid restaking, the launch of EigenDA, and the introduction of novel Actively Validated Services (AVSs) that will utilize restaked ETH. Following Dencun, Ethereum will look to further their discussions on the protocol-level MEV design space, with in-protocol Proposer Builder Separation (PBS) and Protocol Enforced Proposer Commitments (PEPC) as two possible methods of construction. This, alongside Single-Slot Finality, are necessary components of Danksharding and Ethereum’s scaling roadmap endgame, making the Ethereum protocol much more robust if correctly implemented. 

Ethereum is also eyeing ways to decrease state bloat through the introduction of weak statelessness and state expiry, decrease storage requirements through history expiry, and upgrade the efficiency of the underlying architecture by replacing Merkle Patricia Trees with Verkle Trees and incorporating SNARKs across many areas. For more details on these developments, read our report on the Verge and the Purge. It is worth noting that these items have seen less buzz lately due to the emphasis within the industry on scalability at the execution layer, which currently holds priority. 

Arbitrum

On March 23, Arbitrum underwent a paradigm shift with the launch of its native token, ARB, and the establishment of the DAO. Governance over all aspects of the protocol transferred from Offchain Labs, the protocol's original centralized developer, to the broader community of token holders. In an instant, approximately $2 billion in liquid capital was injected into the ecosystem. Around 1.16 billion tokens were distributed to users across over 600,000 addresses, and an additional 113 million tokens were allocated to various protocols. This transition achieved decentralization for the protocol that not many in the space space have been able to achieve. 

The native token holders, now at the helm of the protocol, also control one of the most substantial treasuries in crypto. This treasury comprises about 3.5 billion ARB tokens, valued at ~$4 billion, with ETH revenue regularly flowing in from the sequencer.

So far, the DAO has come to a consensus with respect to two large uses of its enormous treasury: the creation of two grants programs and the short-term incentive program (STIP). The STIP was particularly notable, funding 30 Arbitrum protocols with 50M ARB in total, ultimately allocated as incentives to protocol users. Additionally, a subsequent proposal was recently passed to extend the STIP to an additional 26 protocols to the tune of 21.4M ARB. All incentives for both programs must be utilized by the end of March. Protocols on Arbitrum now have large war chests of their own to encourage the use of their products, which seems to have had a material impact on activity and TVL on the L2.

Two more significant updates came from Offchain Labs with BOLD and Stylus. Today, only 20 whitelisted validators can submit fraud proofs to Ethereum, though the DAO can extend the list. A prospective fraud proof system, named BOLD, is under consideration for implementation by the DAO, which would remove the whitelist and make the system permissionless. BOLD is not yet active on testnet and is likely a bit away, but something to potentially look forward to in 2024. 

Stylus is currently live on testnet and if the DAO decides to implement it on Arbitrum, will allow developers to code smart contracts in Rust, C, and C++ and eventually any language that can compile to WASM. Developers familiar with one of these popular languages may have chosen an alternative L1 but will soon be able to deploy on Arbitrum without additional friction. Stylus will also enable use cases that would not have been possible before, given the prohibitively high gas requirement associated with a more complex Solidity implementation. 

While Arbitrum achieved significant milestones in 2023, it faced competition from Optimism in attracting developers, particularly for building Layer 3s. Throughout the year, Optimism appeared to have a competitive edge in this aspect, drawing more developers to its technology stack to establish rollups.

Optimism

From a development standpoint, Optimism has had a very significant year. Optimism’s greatest achievement of the year was its Bedrock upgrade in June. This contributed to many network improvements, including significantly reduced gas fees resulting from a new method of optimized batch compression, shorter deposit times from the L1 resulting from introduced support for L1 reorgs, and better EVM equivalence and improved node performance as a result of replacing the “one transaction per block” mode with traditional blocks. All of these performance and cost benefits were immediately experienced by network users.

In addition, Bedrock abstracts the proof system from the OP Stack so that a rollup built on top of the technology stack may use either a fault proof or validity proof system and opens the door for the implementation of Cannon for permissionless fraud proofs currently slated to go live in 2024. This is very important, as the current network without fraud proofs cannot reliably inherit Ethereum’s security. By the end of this year, we can also expect another hard fork upgrade called Canyon, which will make Optimism equivalent to the changes deriving from the Shapella upgrade.

Despite these many upgrades and developments, Optimism’s TVL has been mostly sideways, with some momentum towards the latter part of the year. This is mostly in line with the rest of the broader Ethereum ecosystem.

While Optimism’s primary chain hasn’t seen TVL growth, where Optimism has made heavy strides is in its proliferation of the OP Stack, the project’s open-source standard for creating custom L2s. In particular, some large projects have launched or begun development of their own OP Stack chain. Leveraging this tech stack, Optimism is looking to create a Superchain: many OP Stack chains with a set of agreed upon standards that open the door for direct interoperability and collaboration. Optimism introduced the Law of Chains, the first set of standards chains under the Superchain must agree to. This is the first step to creating and fostering the Superchain ecosystem they desire, with the next large step being creating a shared decentralized sequencer set.

The largest launch among the OP Stack Chains came from Base, an L2 created and maintained by Coinbase. With over 7M actively transacting users, Coinbase created Base as a way to facilitate more onchain use cases, and to provide a seamless experience for their users to interact with this onchain world. Base publicly launched on August 9th, but this was preceded by onchain sleuths discovering the L1 bridge contract, sending funds over, and launching meme coins. As a result, Base’s DEX volume surpassed that of all chains other than Ethereum and BNB Smart Chain on July 30 and 31, before the chain was officially live. 

In mid-August, we saw an additional rise in activity due to the launch of Friend.tech, a new social platform that aims to enable creators to connect with their audiences via tokenized attention. This was done by allowing users to buy “keys” of X accounts that have onboarded to the application, giving users access to private in-app chats. Another highly anticipated launch was that of Aerodrome, a fork of Velodrome built and managed by the same team, that quickly became the primary DEX. While activity has largely died down since then, Base has accrued over 3309 ETH in onchain sequencer profit in under four months. 

Besides Base, the only other chains that have committed in some way to the Superchain are Zora, a creator-based chain centered around allowing artists to share their work with their audiences, Mode Network, an L2 focused on fostering collaboration between developers that utilizes contract revenue-sharing incentives, and the Public Goods Network. Other chains that are using the OP Stack but have not committed to the Superchain yet are Worldcoin, a chain focused on the use of proof of personhood and onchain identity, Aevo, a central limit order book exchange for options and perps, among many others. Rollups as a Service offerings Caldera and Conduit have been a large contributor to the traction, providing out-of-the-box solutions for developers to spin up their own OP Stack Chain. 

ZK Rollups

ZK rollups opened the year in the spotlight with zkSync and Polygon zkEVM gearing up to launch, improving Ethereum scalability and interoperability. While their launches have seen varied success in adoption, ZK Rollup usage has been slow out of the gates. We have not yet seen meaningful net-new apps or blue chip DeFi protocols launch on ZK Rollups, while optimistic rollups have successfully attracted both.

Attracting blue chip DeFi protocols starts with DEX liquidity since it plays a key role to scaling lending markets. Uniswap, Curve, and other advanced AMMs need sufficient developer tooling to deploy onto new chains since each ZK Rollup has varying levels of EVM compatibility. If there is a development lift, the new deployment is less likely to happen. Since transaction fees are not sub-cent on Ethereum L2s, orderbook DEXs are not realistic. Further to the DeFi struggles, Chainlink is the default oracle provider for many large lending protocols. Without a Chainlink integration, lending markets cannot easily deploy to a new chain. It's worth watching for Chainlink integrations to trigger a growth catalyst for these ecosystems. 

Richer interoperability is the most promising aspect of ZK Rollups. The solutions enable fast, async communication between chains that cannot by achieved by optimistic rollups. The idea is that chains sharing the same prover and bridge contract can agree to trust each other and quickly pass messages between each other. zkSync and Polygon refer to their solutions as the Hyperbridge and LXLY, respectively. Both ecosystems have teams actively recuriting teams to join their rollup stacks. Gravity, a perpetual futures DEX constructed as an appchain settling down to zkSync Era, is set to be the first new rollup in zkSync’s ecosystem. Polygon is actively collaborating with Canto, OKX, and others to build rollups using the Polygon CDK. If either ecosystem wins the next wave of rollups, it will benefit from the network effects created by improved interoperability.

zkSync

Matter Labs successfully brought zkSync Era to mainnet in March 2023, and the chain has seen strong user metrics throughout the year. zkSync leads the ZK Rollups in bridge contract deposits and DeFi TVL at $493M and $149M, respectively. It executes about 4.7M transactions per day  from roughly 1M unique active addresses at an average transaction fee of $0.31. While the usage metrics are strong, it cannot be ignored that zkSync is the center of attention for airdrop farmers. While no firm plans have been released, the community believes there will be a token that plays a role in the decentralization of the sequencer and the prover. Airdrop farmers believe they will see a meaningful reward for their onchain activity, given Matter Labs has raised $458M for zkSync. 

The ratio of new active addresses (addresses that used the chain for the first time in the period) to total active addresses fell to just 5%, suggesting zkSync is failing to attract new users to its ecosystem. For comparison, Arbitrum, Optimism, and Base see roughly 20% of new addresses in total active addresses. To us, this highlights the absence of new or desired applications that onboard users. This is further substantiated by the top eight DeFi applications by TVL as they are all DEXs. 

Nonetheless, the significant onchain activity has seen over 28.7k ETH ($64.6M) of transaction fee revenue for zkSync. After considering the cost of posting data and validity proofs to Ethereum, the chain generated 11.2k ETH ($25.2M) in gross profit since launch. All system profit accrues to a multisig wallet on Ethereum mainnet that is presumably controlled by Matter Labs given there is no DAO yet and Matter Labs is the sole owner and operator of both the sequencer and the prover. 

Matter Labs just launched its new proving system dubbed Boojum, proving their commitment to continued innovation on the tech stack. The system improves prover performance and reduces hardware requirements, which cuts proving costs and makes prover decentralization more feasible. Boojum is key to the success of the ZK Stack as proofs can be generated and verified cheaply and quickly, enabling fast interoperability between Hyperchains. Boojum has been running on mainnet in shadow mode since July, so the upgrade closes 2023 with an exclamation point for zkSync.

Polygon zkEVM

The Polygon zkEVM also hit mainnet in March of this year but has struggled to attract significant capital into its ecosystem. The new ZK Rollup has $106M of bridge contract deposits and $20M of DeFi TVL. At ~315k transactions per week, the Polygon zkEVM usage metrics are much lower than zkSync.

Without the allure of an airdrop, Polygon’s zkEVM must attract users with quality applications and low fees. Balancer is the only blue chip DeFi protocol that is currently deployed, but Spark protocol is set to launch with DAI, ETH, and wstETH support as recommend by the final risk assessment. With a strong DEX and lending market, DeFi can begin to flourish on Polygon’s zkEVM.

Zooming out, Polygon as a whole is transitioning to Polygon 2.0 with the goal of centering itself around its new zk-powered tech stack. The Polygon 2.0 vision upgrades the protocol architecture, tokenomics, and governance structure to enable. Notably, the Polygon PoS chain will migrate to a zkEVM validium with off-chain DA secured by the existing PoS validator set. Since the zkEVM validium will use the same prover as the existing Polygon zkEVM, the costs to post proofs on Ethereum mainnet is amortized across both chains. As more chains join the ecosystem, the cost should continue to decrease on a per-chain basis. Further to this, Polygon appears to have momentum in the race to attract new chains to join their ecosystem. \

The MATIC token will undergo a 1:1 migration to POL as part of the Polygon 2.0 upgrade. POL will become the broader protocol’s staking token, reward token, and governance token. Most notably, the supply hard cap of 10B tokens will be removed to allow staking rewards to exist beyond the five year window originally provided by MATIC. Emissions are capped at 1% annually and are realistically a requirement for a staking token. POL will likely play a role in decentralizing the sequencer, the prover, or both as the ecosystem matures, giving the token more utility than before. The bridge contract was launched in October and currently 0.4% of MATIC tokens have migrated to POL. However, the bridge is still bidirectional so POL tokens can be bridged back to MATIC. The contract will likely be upgraded to a one way bridge pending a governance vote to begin the official migration. 

MEV

The MEV space had an interesting year. If there’s one trend we are beginning to see, it's a centralization of forces. To begin, roughly 75% of blocks within the past month have been created by the same three block builders, with another 10% going to vanilla builders, showing the immense power distribution happening at the builder level.

This centralization can mainly be attributed to one notable driver:. the growth of searcher-builders who carry out both roles simultaneously, either directly integrating a searcher or partnering with market makers that carry out unique CEX-DEX arbitrage that no other searcher or builder has access to. Examples of these searcher-builders include rsync-builder who have their own order flow as well a partnership with Wintermute, and beaverbuild who have their own order flow and team up Symbolic Capital Partners. The direct partnerships can be seen from the distribution with which searchers send their bids to builders. 

The relay market has also seen a level of centralization, due to the lack of direct economic incentives. As a result of this, Blocknative, a leader in Ethereum infrastructure, decided to shut down their own relay. Prior to shutting down, Bloxroute was among six major relays receiving a significant share of blocks, which have now dwindled down to five. We can expect that without any economic incentive, this number will either further consolidate or be replaced by another part of the stack like block builders who have some incentive themselves. The big trend in this area, spearheaded by the Ultra Sound Relay, is that of optimistic relays, where relays optimistically send the block to the proposer without simulating the block as a way to improve latency. While Ultra Sound and Agnostic Gnosis are the only relays currently using this method, the latency advantage will likely entice other players to adopt this method as well.

Another trend in this direction is that of private order flow, which is a set of transactions that certain entities have access to that others do not. Typically, these transactions are sent directly to a builder based on some agreed-upon partnership. While most of these deals remain private, some of the private order flow can be explicitly tracked. For example, the most prolific MEV-related account this year has been jaredfromsubway.eth, who, at one point, was taking up a majority of blocks on any given day with the strategy of market making and sandwiching traders of long-tail assets like PEPE. Jared has made deals with the four largest builders and sends all of his order flow directly to them as opposed to sending his transactions to the mempool. 

Other important developments within this space include the proliferation of Private RPC services that either mitigate MEV or return some value back to the user. MEV Blocker, introduced by CowSwap, creates an order-flow auction where searchers bid for the right to backrun a trade, allowing for no opportunity to frontrun, which would result in worse execution for end users. 90% of the profit derived from that backrun will be given back to MEV Blocker and finally distributed to end users as rebates. MEV-Share, introduced by Flashbots, has users select what data they share with searchers, who bid for the right to include that data in their bundles. Since searchers cannot see the full value of the transaction, they create partial bundles that may or may not be filled, and any successful bundle has a 90% revenue share with the MEV-Share user. 

While the MEV-Boost architecture continues to become more robust over time, one interesting event this year took advantage of the design. Specifically, malicious proposers were able to take advantage of the MEV-Boost relay design where they falsified the signing of a block header and received the information in the block while actually creating their own. The attack looks like this: the proposer sends out honeypot transactions that lured searchers into creating transactions that they thought would be generating a profit, and as the relay sends the block to the proposer, they would sign a fake block, tricking the relay and including their own block that took advantage of the transactions searchers made. The MEV world continues to be cutthroat, as entities find any way to extract value for themselves, which shows that we may need a little more time before we enshrine this process into the Ethereum protocol. 

Unlike Ethereum’s solution to MEV, the Cosmos ecosystem is taking an alternative approach, led by the Skip Protocol team. Skip is creating a set of modular tools for appchains to implement that give full control over the rules around transaction ordering and inclusion. Not only could these tools create sustainable value accrual mechanisms beyond transaction fees, they prevent the outsourcing of block production to offchain entities. Skip’s vision of the MEV value chain ends with the Protocol itself, while historically, the Ethereum MEV value chain has ended with the validator.

The first implementation of Skip’s work in action was Osmosis’ ProtoRev module, which allows Osmosis to be its own arbitrageur where it receives preference over other searchers in specific backrunning scenarios. To date, this module has generated over $700K in protocol revenue for Osmosis. Skip also launched the Block SDK, which is a toolkit that allows protocols to easily customize their blockspace by introducing mempool lanes that can segment transactions and enforce validation, ordering, and prioritization based on pre-defined rules for each lane. This SDK can create dedicated blockspace for oracle updates, allow for free transactions for first-time users, create a top-of-block auction for MEV recapture, enable orderflow auctions, and improve fee markets that help with congestion. Recently, the Cosmos Hub passed a proposal to incorporate the SDK to introduce an EIP-1559-like fee mechanism to help with resource pricing and mempool congestion.   

Cosmos

For an ecosystem built on the thesis of sovereign, interoperable appchains, its quite hard to summarize everything that has transpired in the Cosmos throughout 2023. In this section, we mainly focus on the developments of the Cosmos Hub and briefly touch on specific appchains related to those developments. We cover Cosmos appchains like Axelar in the interoperability section below and will cover chains like Osmosis, Akash, dydx, and Stride more deeply in the upcoming “protocol highlights” year end report.

Interchain Security Launch

This year was the Cosmos Hub’s most transformative in its entire history. While this may not be saying much, given the fact that it has lacked any utility since mainnet launch in 2019, the Hub finally launched the long-awaited initial version of Interchain Security, called Replicated Security. With Replicated Security, Cosmos chains that want to leverage the Cosmos Hub’s security must be voted in by governance and offer revenue share agreements in the form of tx fees, MEV, and token inflation to subsidize the cost for validators running additional hardware and stakers putting up their ATOM tokens as slashable collateral as a deterrent for malicious activity.

Neutron, a CosmWasm smart contract platform incubated by the P2P team, and Stride, the dominant liquid staking provider in the Cosmos ecosystem, are the Cosmos Hub’s first two consumer chains. Although it is early in the Cosmos Hub’s story as a security provider, the economics behind shared security still need to be ironed out. With 180 validators in the active set, if you assume ~$600/mo per validator to run a Cosmos chain, consumer chains need to distribute at least $1.3M in annual revenue to Hub validators alone. Unfortunately, the value accrual has been abysmal mostly due to the lack of economic activity on Neutron and the slow adoption of liquid staked assets in the Cosmos ecosystem. 

Since early August, the Cosmos Hub has generated a meager $85k from both Stride and Neutron, with Stride contributing over $82k. On the bright side, as part of Neutron’s revenue sharing agreement, it transferred all NTRN unclaimed from the airdrop to the Cosmos Hub community pool, which currently amounts to ~43M NTRN or ~$32M at current prices. While Cosmos Hub validators have been running Neutron at a loss and want to use this NTRN to subsidize their operations, there have been talks within the community to use the NTRN as protocol-owned-liquidity to generate additional revenue for the Hub and help diversify its treasury. 

As the shared security marketplace becomes more competitive with the launch of Eigenlayer and Celestia, the Cosmos Hub needs to find a moat and identify additional ways to capture value. Shared security alone is likely to become a commoditized offering and may never become a sustainable revenue driver as transaction and MEV fees trend towards zero. Even though Interchain Security was supposed to be a catalyst, there is a growing realization by the community that chains do not want to pay for the full security of the Cosmos Hub this early on in their lifecycle and that ICSv2, protocol-owned-liquidity, and ATOM as a preferred form of collateral in the Cosmos ecosystem is the future of the Cosmos Hub.  

Updated Cosmos Hub Tokenomics

At the end of 2022, a team of prominent Cosmos community members presented a whitepaper to the community that outlined a vision for the Cosmos Hub. A signaling proposal to approve of these plans was shot down in a contentious vote due to concerns with the drastic changes in ATOM’s monetary policy, including a large front loaded ATOM mint that would go into a treasury without any explicit governance oversight.

In mid-2023, a community grants team called the ATOM Accelerator DAO funded three teams to reimagine the Cosmos Hub’s tokenomics and governance in the era of Interchain Security. The two major results of these teams’ work concluded that the Cosmos Hub was overpaying for security, with ATOM having the highest issuance rate out of all major L1s, and introduced the idea of a Dynamic Liquid Staking tax. The dynamic liquid staking tax is a way to align the monetary and fiscal policy of the Cosmos Hub with the growth of liquid staking and could become a substantial value capture mechanism in the form of an ATOM burn or as a means to strategically increase the Cosmos Hub’s treasury for protocol-owned-liquidity deployment. Additional ideas were proposed such as a Vote Power tax to help with the poor distribution of stake of the Cosmos Hub, the ability for the Cosmos Hub community to influence governance of other protocols in the ecosystem and use the Hub treasury for protocol-owned-liquidity, and the establishment of a developer incentive fund to fund teams that can build revenue-generating products owned by the Cosmos Hub.

The first proposal influenced by these findings was Proposal 848, which sought to reduce ATOM’s max inflation parameter from 20% to 10%, and at the time of passing reduced ATOM’s actual inflation from ~14.3% to 10%. This proposal had the highest voter turnout in the Hub’s history and was extremely contentious with many, including Cosmos Founder Jae Kwon, believing the Cosmos Hub would become less secure and lose its position as the premier security provider of the ecosystem. With the passing of proposal 848, Jae has decided to launch a Cosmos Hub fork called “ATOM1” that plans to keep the original monetary policy and compete with the Cosmos Hub. 

In the coming months, there will be subsequent proposals to lower the minimum inflation of the Cosmos Hub, increase the inflation change parameter, and put up signaling proposals to receive community approval to fund development related to the tokenomics research. Overall, 2024 stands to become a “make or break” year for the Cosmos Hub even though new chains like dYdX and Celestia are bringing awareness and liquidity to the Cosmos ecosystem.

Noble Brings Native USDC to the Cosmos

The collapse of TerraLuna created a huge gap in the Cosmos ecosystem for stablecoin liquidity. Post-collapse, the ecosystem relied on a bridged version of USDC that increased the barrier of entry for capital without any major fiat onramps and limited DeFi growth. Additionally, a prerequisite for dYdX to successfully migrate from Starknet to Cosmos was the deployment of native USDC. In 2023, Noble, a general asset issuance appchain, brought native USDC to the Cosmos ecosystem and has seen impressive growth in a short period with dYdX chain being a major source of USDC demand. As of December 10, 2023, Noble USDC has grown to over $40M, surpassing circulating native USDC in ecosystems like Near and in-line with native USDC issued on Optimism. With over $350M TVL in dydx’s Starkware deployment, one should expect Noble USDC to continue to grow in 2024.

dYdX V4 showcased a unique CCTP integration by plugging it directly into the v4 appchain and obfuscating bridging for the user. Since CCTP launched on Noble at the end of November, there has been $8.8M USDC transferred into Noble and $4.7M out of Noble as of Dec 10, 2023. The launch of CCTP has led to increased USDC liquidity and activity across the Cosmos ecosystem, with total USDC IBC volume steadily increasing as shown by the chart below. Tracking USDC inflows into Noble and subsequent outflows to specific appchains in Cosmos should be an indicator to watch closely.  

Lastly, Noble was supposed to be a big catalyst for the Cosmos Hub, as the third planned consumer chain to leverage its security. In early December, Noble announced the delay of its ICS launch due to economic concerns around Replicated Security. Currently, Noble doesn’t believe it can economically sustain all 180 validators of the Cosmos Hub and is waiting for ICSv2 to launch in 2024, which allows validators to opt into securing a chain if they so choose.

Celestia Launch

Outside of native USDC deployment and the migration of dydx into the Cosmos, the major story for the wider Cosmos ecosystem was the launch of Celestia in Q4. Although the modular thesis continues to play out on Ethereum, Celestia offers much cheaper data availability for rollups and has already seen promising projects like Eclipse, Movement, and Manta Pacific announce plans to leverage it over Ethereum. While it remains to be seen if network effects can coalesce around a data availability layer, the crypto space has yet to see a DA-specific project launch with a liquid asset, so speculation for TIA will likely continue throughout 2024 and will create secular tailwinds for Osmosis as the major onchain venue for TIA trading.

Solana

In the wake of the FTX collapse, 2023 proved to be an exciting year for the Solana ecosystem as market participants slowly began to understand the vision and utility for a monolithic, high performance, and low fee L1. Being tied to the hip with Alameda and FTX, SOL experienced heavy sell pressure going into 2023. Serum, Solana’s native orderbook, which powered essentially all DeFi activity on the network, was deprecated due to program update authorities residing in defunct FTX wallets. Our report from February more deeply covers the contagion of these market dynamics. Despite these hurdles, Solana experienced a redemption arc, primarily through UX improvements, a flourishing ecosystem of dApps, and user onboarding. 

A Refined UX

Solana Mobile, a subsidiary of Solana Labs, opened sales for Saga, an Android-based smartphone optimized for Solana, in April of 2023. To ramp up device sales, the company reduced the price from $1,000 to $599 in August. Saga is an ambitious product vertical to execute in the crypto industry. Alongside the launch of Saga, Solana Lab bootstrapped a $10M grant program to incubate promising dApps for the Solana Mobile Stack (SMS). While most of the crypto UX happens on PCs, Solana Labs bet big on a mobile experience. Other industry titans, such as Coinbase, also went for this same approach with Coinbase Wallet and in-app browser capabilities. 

Another major UX unlock was seen in the launch of Backpack, a wallet that supports executable NFTs (xNFTs), which essentially serve as tokenized in-wallet dApps. For example, a user could lend or borrow funds on MarginFi’s money market directly from within Backpack. This serves as a major innovation in the crypto wallet design space, where for the first time a wallet can be seen as more of an operating system than a browser add-on. To top it off, Backpack announced that it will launch Backpack Exchange, a centralized exchange, for the public in the first quarter of 2024. While initially it will only offer spot crypto markets, the team plans to introduce derivatives and margin trading later in the roadmap. The exchange plans to utilize zk proofs and multi-party computation (MPC) to secure customer assets. 

Ecosystem Development

After a complete shakeout in the DeFi sector, Solana saw several dApps rise from the ashes and continue to ship new features for users. The list below is by no means all encompassing, but instead shows some of the dApps that generated the most buzz throughout the year.

The ecosystem picked up significant momentum during Solana’s Breakpoint conference in November, which coincided perfectly with a broader market rally. Several airdrops were announced around Breakpoint, including Jito, Jupiter, and Pyth. This led to an influx of new entrants on the network, who were eager to interact with major DeFi protocols in an effort to earn retroactive token rewards. In turn, SOL price appreciation plus potential airdrops ignited a flywheel in the Solana ecosystem, leading to a surge in stablecoin transfer volume, daily users, and TVL. 

At the end of 2022, Jito Labs launched its own liquid staking derivative, jitoSOL, which enhances users’ yield by passing along MEV tips in addition to staking rewards. Further, jitoSOL is composable with DeFi, and can be used as collateral across lending dApps. Over the year, jitoSOL has grown to 6.6M SOL ($396M), which comes close to the incumbent leader Marinade Finance in the domination of SOL liquid staking. Further, Lido voted to discontinue SOL support, creating room for Solana-native protocols to win market share. Given that liquid staking on Solana only comprises ~4% of total staked SOL, both Jito and Marinade stand to benefit as more participants look to stake and earn SOL. As liquid staked SOL supply grows, both protocols are likely to earn fees and pass revenue to token holders. 

DeFi aside, one vector that Solana has had continued success in is NFTs. NFT marketplace Tensor, tailored for NFT traders and comparable to Ethereum’s Blur, saw a breakout year in volume, thanks to its reward program and several hot NFT collections. Backpack’s MadLads NFT mint in April kicked off a wave of volume that saw a second wave in the last quarter of 2023. Other than speculative profile pictures, Solana also saw an innovation in state compression, a new mechanism that enabled the minting of compressed NFTs (cNFTs) with drastically lower storage costs than ever before. Web3 messaging dApp Dialect utilizes cNFTs to mint millions of NFTs for fractions of a dollar, letting users collect unique images and stickers from their favorite creators. 

One of the biggest challenges for Solana in 2021 and 2022 was network performance. The network could not support transaction demand as more users entered the ecosystem, evidenced by six different periods of network performance degradation in 2022. However, several improvements implemented towards the end of 2022 provided a clean slate for unfaltering uptime in 2023. Validators transitioned from User Datagram Protocol (UDP) to QUIC, a general-purpose protocol used to relay transaction requests that dynamically adjusts to request overflow. The network also implemented local fee markets, which introduced priority fees, which allows users to pay surplus fees on top of a base fee to prioritize a transaction. 50% of Solana’s priority fees are burnt. Both QUIC and local fee markets contribute to Solana’s parallel processing capability, enabling the network to handle large transaction loads on certain programs without degrading overall performance. For a more in depth discussion of these technical updates, be sure to read our previous report. 

2023 brought a new token standard, Token-2022, to Solana, which opened the development landscape to a new realm of possibilities. Prior to Token-2022, most tokens on Solana were SPL tokens, which are not necessarily 1:1 with Ethereum’s ERC-20 token standard. In response, the Solana Foundation introduced Token-2022, which enables several new features, such as rebasing, soulbound tokens, and anonymous transfers. This should lead to further innovation in Solana DeFi going forward. Perhaps the most exciting ongoing development for Solana is Jump Crypto’s upcoming Firedancer validator client, which aims to significantly ramp up throughput capabilities while enhancing validator decentralization. Firedancer will likely go live in 2024 and mark a new era of monolithic Layer 1s.

Move L1s

After Meta’s (formerly Facebook) Libra blockchain shut down, the team parted ways to start two new alt L1s, Aptos and Sui. Both of these networks rely on a new programming language purpose-built for blockchains - Move. In short, Move builds on the same core technology that powered Libra. It utilizes a formal verifier for smart contracts that provides additional safeguards against malicious behavior, which in turn enables flexible private key management and hybrid custodial options to optimize asset security for users. Move’s architecture also helps prevent reentrancy attacks and optimizes memory management. With around $3B in hacks over the years, there is certainly a need for a safer smart contract development language, but the merit of Move’s security is debatable until it is more widely adopted. These newer networks are monolithic in nature and have the same value proposition as other alt L1s: fast, low-fee transactions and vertical scalability. While the technology does offer these benefits, an entirely new smart contract language and wallet experience have likely caused some friction which has limited adoption.

Aptos, being the first Move-based blockchain to launch on mainnet, has not seen much adoption since its late 2022 release, especially when compared to other L1s and L2s. However, the network unveiled a couple new developments throughout 2023, such as a gaming SDK and Quorum Store, which aims to eliminate bottlenecks in consensus. Aptos Labs plans to unveil a new Move compiler to improve the developer experience. This will likely be unveiled in the first half of 2024. Despite the progress on developer tools and network performance, DeFi activity on Aptos remained relatively stagnant throughout the year. In 2024, the network will need to tackle one of the hardest obstacles in crypto, the cold start problem. Desirable dApps, whether they involve DeFi, gaming, or AI, will need to surface for adoption to happen. 

Sui experienced exponential growth since its launch in Q2 of 2023, with TVL increasing 500%. The tail end of this growth coincides with Sui Foundation’s 117M SUI ($50M) ecosystem fund, which catalyzed a boost in developer activity. However, it is difficult to judge the merit of this growth, given that most of the protocols live on mainnet are copies of each other. For example, there are 10 different DEXs, three lending protocols, and two derivatives protocols. Sui experienced a surge in transactions in July, thanks to Sui 8192, its first onchain game. While this serves as a proof of concept for how the network handles periods of high network demand, Sui continues to evade center stage in the shadow of Solana and L2s. With that being said, the network shows promise as a functional alt L1, primed for more user activity in the future. 

Near Protocol

2023 proved to be a breakout year for Near. Although Near’s DeFi activity continued to slow down, the protocol saw three key developments, including a surge in network usage through consumer adoption, a boost in traction on its Blockchain Operating System (BOS) product, and a strategic pivot to becoming a data availability layer. Near also implemented several updates throughout the first half of the year to improve the developer and user experience. One update is meta transaction support, which is a feature that can be most easily compared to account abstraction on Ethereum. Another key development was the implementation of FastAuth, a new onboarding mechanism where prospective users can set up a wallet by using their email and phone prompts, such as iPhone’s Face ID or a passcode. This abstracts away the need to manage seed phrases or long strings of numbers from the user while strengthening authentication security. These developments were key to simplifying the onboarding process for adoption of popular dApps.

Near’s two most popular dApps are Sweatcoin and KaiKai. Sweatcoin was Near’s first consumer-facing dApp, and it incentivizes users to track walking/running activity to receive SWEAT tokens, which can be redeemed for goods and services in Sweatcoin’s marketplace. In April, Near partnered with Cosmose AI, an AI analytics company that tracks store traffic and engages with shoppers to create a more efficient payment system, granting both buyers and vendors savings on transaction fees. Cosmose owns the KaiKai app, which creates Near wallets on behalf of users. KaiKai users can use fiat to purchase KAIC, a stablecoin pegged to $0.01, and use it within the app to receive discounts on purchases. Sweatcoin and KaiKai led to a steady increase in daily active users and transactions, and it is likely that Near will continue to foster more creative consumer dApps in 2024. This goes to show that Near’s strategy differs from a lot of other alt L1s. Instead of recreating the same AMMs, derivatives, and lending markets found on Ethereum, the protocol focused on creating dApps that are attractive to consumers. Network growth for Near does not need to be reliant on DeFi volume. 

Near also introduced the Blockchain Operating System (BoS), an open source operating system that simplifies the complexity of the Web3 experience. It can best be compared to Shopify, but for crypto. In doing so, the BoS makes it easier for developers to build on any chain by using familiar languages and easily forkable modules to rapidly create and distribute dApps. Developers gain the opportunity to quickly ship products without worrying about blockchain-specific production hurdles, such as setting up validator infrastructure or bulletproofing a smart contract. In November, Near announced its data availability layer, which will help developers reduce costs while maintaining security with Ethereum. Despite being out of the spotlight for some time, Near continued to ship new product offerings and build network traction throughout 2023, and this momentum will likely continue throughout 2024. For a more in depth look at Near’s 2023 progress, be sure to read our report. 

Avalanche

2023 was a year of building for Avalanche with several notable developments. The introduction of the HyperSDK significantly enhances developer tooling by enabling application-specific chain customization at the virtual machine level. This means developers can tailor a subnet’s VM to their specific function, enabling novel use cases and optimizing end user experience. Another new tool, Avalanche Warp Messaging (AWM), will soon allow seamless communication across Avalanche subnets. AWM removes the need for third-party bridges to transfer data or assets between chains in the ecosystem. These two developments significantly increased Avalanche’s appeal to developers looking to launch app-specific blockchains, creating a more formidable competitor to other technologies such as the Cosmos SDK.

Throughout the year Avalanche proved itself as a top tier chain in the realm of gaming. Games like Crabada, DeFi Kingdom, Beam, Shrapnel, and Gunzilla created their own custom subnets. Other games like Domi Online, Off The Grid, and TSM Blitz are coming soon. The newly launched Avalanche Arcad3 aims to connect established gaming giants and esports organizations with Web3 game studios. By attracting major partners and developers as well as allocating significant resources, Avalanche has cemented itself as a major competitor in the future of GameFi—a growing narrative. 

Additionally, Avalanche caught the attention of TradFi, particularly with its custom Evergreen subnet framework. Evergreen subnets are permissioned and geared towards institutions with turnkey compatibility with financial regulations, such as geoblocking and KYC. Specifically, the first implementation of Evergreen, dubbed Spruce, was supported by T. Rowe Price, WisdomTree, Wellington Management, and Cumberland. Additionally, J.P. Morgan has shown interest in the technology. Despite these substantial developments, the overall network adoption of subnets has been lackluster to date, though in 2023 the number of subnets increased from 23 to 51. 

A new governance proposal, ACP-13, looks to completely overhaul AVAX’s tokenomics and the tech stack’s appeal among developers. Though it is early in its life cycle and far from being implemented, ACP-13 would remove the requirement for a subnet validator also to validate the primary network. Today, in order to become a validator on a subnet, the validator must additionally be actively staking 2,000 AVAX. The proposal looks to remove this requirement and exchange it with an ongoing variable fee dependent on the usage of the subnet. If the proposal gains traction, it will substantially change the fundamental value proposition for AVAX and likely the appeal of creating a subnet for developers. 

Avalanche experienced fluctuations in its market presence and value. Its TVL market share across chains decreased from around 2% to approximately 1.7%. The AVAX token spent most of 2023 declining against its peers but had a strong finish at the end of the year. This recovery has restored AVAX's value against ETH to a level above where it started from at the beginning of the year. 

Avalanche showcased impressive retention rates against other L1s. 'Retention' in this context is defined as the percentage of addresses that continually engage in transactions each month following their initial use. The activity from users signals the network's ability to maintain an active and engaged user base over time.

Oracles

The oracle vertical is inherently tied to the DeFi market, and as activity in the latter sector rebounded at the end of 2023, it’s no surprise that oracle project KPIs followed suit. The only oracle provider that experienced notable growth throughout 2023 in terms of total value secured (“TVS”) is WINkLink, going from ~$3.5B at the beginning of the year to ~$7.5B as of December 13. However, WINkLink—a Chainlink fork—solely operates within the TRON ecosystem, a market that generally isn’t pursued by any other provider, and only offers services to two protocols.

Another project with a limited set of customers is Chronicle, which was spun up by MakerDAO’s Oracle Core Unit. The project has historically been exclusively utilized by MakerDAO but expanded to public availability in September. As it currently stands, Chronicle is only leveraged by Keep Network in addition to the MakerDAO ecosystem, and the project focuses on expanding to a wider set of blockchains. Chronicle launched through a community-enabled network of 22 node operators who provide data observations for prices on six different asset pairs on Ethereum, nine asset pairs on Polygon zkEVM, and four asset pairs on Gnosis Chain. The project implements Scribe, a product leveraging an optimistic Schnorr hybrid mechanism, which compresses data from multiple oracle node operators into a single signature. Verification of such signatures is expensive and slow on low throughput blockchains due to heavy computational requirements, which is why data updates are optimistic on Ethereum, with third parties being able to dispute updates onchain. If a data update isn’t successfully disputed within a predefined time period, the update will be considered finalized. 

The number of customers an oracle protocol serves can be argued to be a more meaningful performance metric than TVS since it better showcases the provider’s ability to monetize its services. The three notable projects with diversified customer bases are Chainlink, Pyth, and Uniswap TWAP, although TWAP shouldn’t be thought of as a direct competitor to the rest of the oracle protocols as it simply offers access to onchain time-weighted average prices of Uniswap pools. Chainlink has maintained its monopolistic position within the vertical, introducing several new products in 2023. However, overall demand for these products has been lackluster, and Pyth has emerged as a notable competitor by creating a robust market position within low-latency oracles.

Chainlink

During the year, Chainlink launched two new oracle products, Data Streams and Functions, in addition to its cross-chain messaging solution, Cross-Chain Interoperability Protocol, which will be covered in detail further below. Data Streams is the project’s response to somewhat of a trend within the oracle vertical, where protocols are increasingly focusing on low-latency price data updates through pull-based oracles. This mechanism is particularly valuable for perp exchange protocols to mitigate oracle extractable value. Unlike Chainlink’s main product, Data Feeds, which updates onchain data through a reference contract once a value deviates beyond a specific threshold or a certain time has passed, Data Streams oracle nodes continuously sign and deliver aggregated data offchain, from where users can pull the data onchain on demand.

The product is in Mainnet Early Access, and consequently, Data Streams hasn’t yet seen a significant amount of usage. It should be mentioned that GMX was the first protocol to implement the product together with Chainlink’s Automation service. This was a major win for Chainlink since, in return for the tailor-made product set, GMX entered into a fee-sharing agreement to allocate 1.2% of its V2 and later protocol versions revenue to Chainlink. Such a revenue model should be significantly more beneficial for Chainlink than the per-data update payment method utilized for Data Feeds, as in the latter model, customers share the update fees between themselves. If Chainlink can push the fee-sharing model more aggressively in 2024, the project could significantly enhance its value capture. 

The Functions product is currently in Beta and was deployed on Ethereum, Avalanche, and Polygon Mainnets at the beginning of October 2023. The service grants users access to custom compute on data, AI models, cloud services, legacy TradFi systems, etc. Moreover, in 2023, Chainlink provided updates on products that have been under development for some time and could be released during the next year. On the data side, more information on DECO was provided, which is a privacy-preserving oracle protocol and could be leveraged for use cases such as identity verification and undercollateralized lending. Through zero-knowledge proofs, DECO will enable data sources utilized by oracles to prove the correctness of their data without revealing the data itself. Chainlink is also working on a transaction ordering solution called Fair Sequencing Services (“FSS”), but this product seems to be in its early development phase. Nevertheless, the goal of FSS is to mitigate malicious MEV by using Chainlink’s oracle networks to order transactions fairly. 

Put differently, Chainlink expanded further away from its core offering in 2023, which is set to continue in 2024, signaling that the project is beginning to focus more aggressively on new demand sources. Spreading more thinly into new market sections creates a first-mover advantage for Chainlink but has the downside of potentially eroding its current strength with respect to price data oracles. This is something to monitor throughout 2024.

Pyth

During the year, Pyth’s low latency oracles gained significant adoption, both within and outside the Solana ecosystem. At the start of the year, the total number of integrations stood at 92, with more than 50% of the integrations on Solana. By year-end, the total number of integrations has reached 270+, with less than 30% of those on Solana. 

The Perseus Network Upgrade in September 2023 was a significant improvement in the oracle network’s architecture. Previously, the attester program on Pythnet batched 5 price updates into each message, meaning that consumers had to update other unneeded prices as long as they were in the same batch, which also resulted in consumers having to pull multiple batches just to update the price feeds they needed. Now, consumers can group together any set of feeds into a single payload, containing exactly the prices they want, significantly saving on gas costs.

Another significant driver of growth was the price scheduler. In July 2023, Pyth entered into a strategic partnership with Aptos to deliver its price feeds through an automated price scheduler. This means that the Aptos Foundation helps pull price updates onchain at a sub-second frequency, and any dApps building on Aptos can access those price feeds at no cost. A similar service also became available on Sui. 

Oracle Value Capture

Historically, the standard operational model for oracle networks has consisted of having sets of permissioned nodes that don’t have any explicit economic disincentives for misbehaving due to node operators not being required to stake tokens to join a network of nodes. Instead, node operators are generally well-known entities, with oracle projects trusting these operators not to act maliciously based on potential reputational damage as well as loss of future profits.

The aforementioned is starting to change, though, with Chainlink and Pyth both slowly introducing a staking mechanism to their respective oracle network designs. In December 2022, Chainlink implemented Staking v0.1, an early version of the project’s vision for staking, enabling 25M LINK to be staked with node operators supporting the ETH/USD Data Feed on Ethereum. This version was very rudimentary, e.g., stake slashing wasn’t enabled.

Beginning in late August 2023, Chainlink began promoting v0.2 of staking, which was later introduced in the form of a three-tiered migration scheme throughout late November and early December. Staking v0.2 enlarged the staking pool to 45M LINK and introduced node operator stake slashing, with the purpose of securing more services in addition to the ETH/USD Data Feed on Ethereum. As an increased number of services become secured by staking, the ecosystem’s cryptoeconomic security will start growing. However, it should be noted that staking yield continues to be based on a fixed reward allocation and is presently fully paid out through inflation rather than node operator gross revenue.

When it comes to Pyth, the project released its V2 whitepaper in September 2023, which included information about a PYTH token. The total token supply is 10B, with 85% of the tokens initially locked. Following the release of the whitepaper, in November 2023, Pyth announced a retrospective airdrop, which airdropped 6% of the supply to DeFi participants who have used Pyth-powered dApps, community members, and protocols that have integrated with Pyth. 

As the PYTH token is live, staking will be enabled once approved through a governance process. Publishers will have to stake PYTH on Solana in order to participate in the Pyth network as a publisher, as determined by governance. If any slashing occurs, the slashing will occur on Solana. In addition, the Pyth network may eventually use a price aggregation function based on stake weight, meaning that publishers with more PYTH staked have a larger weighting in the aggregation program on Pythnet. 

The oracle vertical has long been dominated by Chainlink, with Pyth having managed to capture noteworthy market share since the middle of 2022. Smaller players have emerged from time to time, but none was able to penetrate the market in 2023 meaningfully. Protocols across the board have struggled with value accrual, where the oracle service can become prohibitively expensive for end users, especially for early-stage projects, while node operators enabling the service aren’t compensated adequately relative to the gas fees and operating expenses they incur.

The aforementioned exemplifies a vast, current problem within the oracle vertical: service providers are enabling a massive amount of high-value transactions for which they can’t charge adequately. In general, projects are operating with broken business models that can’t be maintained in perpetuity. For example, with the exception of Chainlink’s Price Feeds service, node operators were heavily subsidized by the project in 2023 such that operators could stay profitable, while Pyth’s operators were working practically for free. To be clear, this isn’t really a problem for major oracle protocols with highly valued tokens in the short to medium term since they can artificially increase node operator profitability by subsidizing them through native token inflation. 

Consequently, it seems increasingly likely that the long tail of the onchain price data market will continue consolidating in 2024, with Chainlink strengthening its position on EVM-chains while Pyth gains traction on alt-L1s. Moreover, the market exhibits monopolistic dynamics since switching costs can be high, protocols enjoy strong network effects, and the Lindy effect plays an important role. Chainlink is uniquely positioned due to the width of its product offering, but the demand simply isn’t there yet in markets adjacent to price data. Therefore, with the exception of Data Feeds and Data Streams, the rest of the project’s oracle and compute products will likely continue as unprofitable in 2024 until DeFi activity returns or new blockchain-based use cases are introduced. 

Bridges and Cross-Chain Messaging

Bridges have had a tumultuous year. As with previous years, bridge hacks were responsible for some of the largest exploits in 2023, with the most damaging being Multichain’s $126M exploit. As shown below, TVL for the five largest bridges has remained flat for the entire year.

Bridging infrastructure was in hot demand when the multichain thesis dominated attention at the top of the last cycle. While that thesis remains true, we have seen consolidation around a few core ecosystems – Solana, Ethereum and its L2s, and the Cosmos. Bridges that best service these routes are best positioned for success. 

Bridge architectures have largely remained stagnant these years. Synapse’s launch of an OP stack chain for its interchain network was largely disappointing. One notable trend in the bridging space is the rise of intents based bridging models, with Across taking the crown in that for the time being. In intents based bridging models, relayers/solvers/fillers front the liquidity on the destination chain while waiting for the underlying bridge settlement to occur so that they can be repaid. Such models often provide a much better bridging experience for users, with users receiving much faster execution and better quotes.

Another trend in the bridging space is the rise of bridge aggregators or bridging middleware solutions. For example, LI.FI is a developer-focused middleware solution that offers a bridging API, helping products integrate a bridging solution into their protocol. The API aggregates liquidity from 15 bridges, 30 DEXs and solver networks, 20 EVM chains, and Solana to offer the best execution for bridging and cross-chain swaps. To date, they have processed $2.0B+ in volume and 2.5M+ transactions. 

A similar competitor, Socket, has built a liquidity meta-layer for cross-chain swaps and bridge transactions, and similarly have achieved integrations in reputable wallets such as Metamask and Coinbase Wallet. To date, Socket has transferred $5B in volume over 2.5M+ transactions. Both LI.FI and Socket have developed their own front ends for their bridge API solutions, respectively Jumper and Bungee. 

LayerZero

LayerZero saw a large burst in user activity in the middle of the year, largely as a result of increased user activity on Stargate. However, it is important to note that Stargate activity in has largely been below 40% of total LayerZero cross-chain messages, with that metric standing at 33% for the month of November. Since the largest burst in activity in the middle of the year, LayerZero messaging activity has significantly decreased as shown by the graph below. In May 2023, Polyhedra rolled out a production-grade ZK light client, built directly on top of LayerZero, which offers another secure oracle solution amongst other LayerZero provided oracle options. In September 2023, LayerZero onboarded Google Cloud as an available verifier network.  Omni-chain fungible tokens (OFTs) have been adopted by protocol after protocol, and LayerZero’s OFT standard has managed to capture a significant amount of mindshare as a cross-chain token standard. Some of the largest OFT adopters include MIM, JOE, BTC.b, and now LayerZero’s own implementation of wstETH.  

On December 14, 2023, LayerZero launched V2 of the protocol. The upgrade was a significant architectural upgrade to the core protocol, addressing previous developer and user concerns around certain security assumptions, such as the oracle + relayer set and execution guarantees. V2 introduces an executor role, which allows the protocol to decouple security + execution, in comparison to V1, where security + execution was intertwined due to the relayer role. The protocol maintains its liveness and censorship resistance guarantees with permissionless executors and lazy nonce order enforcement, while improving its horizontal composability. It remains to be seen whether LayerZero V2 will be enough to reignite protocol growth. 

Circle CCTP

On April 26, 2023, Circle launched the Cross-chain Transfer Protocol to improve USDC bridging between blockchains. The new transfer protocol provides a signed attestation that states a user has burned a certain amount of USDC on the source chain which allows the user to mint the same amount of USDC on the destination chain, removing the need to have bridge liquidity on the destination chain. Initially rolled out on Ethereum and Avalanche, CCTP is now supported on Optimism, Arbitrum, Base and Noble with imminent launches on Solana and Polygon PoS. 

Given CCTP enables low fee transfers with no slippage without needing source destination chain liquidity, it is poised to reshape the bridging sector. Liquidity based bridges now seem bulky and provide poor UX. However, it is important to note that often CCTP has a slower bridging time when compared to other liquidity based bridging models. As of December 10th, 2023, CCTP has processed over $1.07B in bridging volume with 40K unique wallet addresses making cross-chain transactions through CCTP. Both CCTP volume and transaction count have shown significant growth, especially for the month of November, and with Noble just getting started along with imminent Solana and Polygon PoS deployments, we expect the trend to continue. 

The USDC marketcap has struggled during the bear market, falling -57% from $55.9B to a low of $23.8B. CCTP should improve USDC adoption and usage given it will play a key role in the new bridging paradigm. Notably, Tether has yet to release a similar solution. 

Chainlink Cross-Chain Interoperability Protocol

The Cross-Chain Interoperability Protocol (“CCIP”) was the most anticipated product launch by Chainlink this year. It entered a Mainnet Early Access phase on Avalanche, Ethereum, Optimism, and Polygon in the middle of July, with Arbitrum, BNB Chain, and Base Mainnets added since. Unlike traditional bridge protocols that target end users directly, the CCIP business model revolves around usage deriving from integrated protocols and, possibly, financial institutions. To transfer data across blockchains, Chainlink utilizes unidirectional and distinct pathways called Lanes, with each Lane consisting of three separate offchain node networks that must be in agreement for transfers to be finalized. This configuration is built to prioritize security over everything else, which is in contrast to most other projects within the vertical that generally optimize for speed and cost.

Since the launch of CCIP, Chainlink has strived to spin up a highly opportunistic narrative regarding the cross-chain messaging protocol being adopted by financial institutions and becoming the standardized solution for connecting TradFi entities’ legacy systems to the onchain ecosystem. CCIP has managed to set up impressive partnerships with some of the world’s largest financial institutions, the Depository Trust & Clearing Corporation, as well as SWIFT, engaging in test-phase collaborations where CCIP has been leveraged to move and settle tokenized assets between public and private blockchains. All of the aforementioned partnerships are in the early stages, and it’ll likely take several years before financial institutions begin actively incorporating onchain components into their operational models. In addition, there’s a notable risk that TradFi entities form an onchain-focused alliance similar to how SWIFT was originally created, or governments force their way into the vertical through regulation, creating proprietary solutions and pushing Chainlink out of its greatest potential target market.

Despite CCIP having over 100 partnerships with financial institutions and crypto projects, usage was lackluster in 2023. The cross-chain protocol is still in a Mainnet Early Access phase, and it seems as though only two protocols, Aave and Synthetix, have leveraged CCIP in a somewhat meaningful manner. In the middle of 2023, Synthetix announced that it integrated CCIP into its V3 upgrade for cross-chain token transfers. The project also decided to utilize Wormhole for a different cross-chain reads use case. Despite headwinds, it’s worth monitoring CCIP activity in 2024. Mainnet General Availability is poised to launch at the beginning of the year, and partner protocols will likely begin leveraging the solution more actively. The market has been underwhelmed with CCIP usage, with many writing off any short-term potential, which could warrant a new LINK narrative if activity suddenly picks up.

Axelar

Axelar has been the dark horse in the interoperability space with LayerZero and CCIP dominating the air waves despite having more chain connections and the ability to offer asynchronous cross-chain composability with the launch of General Message Passing (GMP) earlier this year. Today, Axelar is the only interoperability solution that allows general cross-chain swaps and DeFi activity between the Cosmos and EVM. Additionally, with applications like Squid Router leveraging GMP to offer one of the best cross-chain swap experiences, Axelar adoption continues to see strong growth. 

In 2023, Axelar (and Wormhole) were chosen by Uniswap governance as the two ideal solutions to facilitate secure protocol updates governing Uniswap’s cross-chain deployments, edging out interoperability solutions like LayerZero. Axelar has also received attention from major traditional financial institutions like JP Morgan and Apollo who developed a PoC under Singapore’s MAS “Project Guardian” that leveraged Axelar to demonstrate how the technology could be used to manage large-scale client portfolios, execute trades and automate portfolio management of tokenized assets. 

With LayerZero’s token launch likely imminent in the first half of 2024, there may be a rerating of AXL to the upside. In addition, Wormhole chain is launching in 2024 and offering the only interoperability solution between Cosmos, Ethereum, and Solana, so Axelar may need to work towards a Solana integration to remain competitive.

Interblockchain Communication

With bespoke bridging solutions all vying to become the standard in the crypto space, liquidity fragmentation and poor cross-chain UX will continue to be a concern as more L2s launch and other ecosystems grow. The need for a trust-minimized bridging standard is sorely needed and no solution is better positioned today than Cosmos’ IBC. 

Today, IBC averages over $25M of daily bridge volume when not double counting inflows and outflows. This would make it a top 3 interoperability solution, in line with Arbtitrum and Polygon’s native asset bridges. What's important to recognize is that IBC is not a simple asset bridge, but instead a general messaging standard that can pass any data packet from one chain to another. 
Below shows daily IBC outflows from the Cosmos Hub and Osmosis and does not represent the total IBC volume referenced above. Historically, these two chains account for ~40% of daily IBC volume. One can see a notable uptick in the combined 30D moving average of IBC outflow volume of these two chains post-Celestia and dYdX v4 launch from ~$4M to ~$10M. Also to note, the large spike in mid-August coincided with the launch of Sei, which didn't have a direct connection to the Cosmos Hub and only took ATOM deposits from Osmosis. This likely caused a large spike in outflows from the Hub to Osmosis and then large outflows out of Osmosis to Sei.

Historically, Cosmos cross-chain UX has been clunky even though all SDK chains have leveraged IBC. In 2023, teams like Informal Systems, Skip protocol, and Strangelove focused on improving IBC relaying infrastructure and abstracting away IBC by creating things like Packet Forward Middleware to help with IBC path unwinding for multi-hop routing and also introducing Skip API, a unified RPC service that enables frontends to offer DEX aggregation over IBC. Additionally, the introduction of Interchain Accounts and Queries allows for asynchronous cross-chain composability over IBC that teams like Stride have leveraged to create a 1-click liquid staking experience that has historically taken 5+ transaction approvals. While L2s, particularly Optimism, are promoting the Superchain vision without an interoperability standard, many will be surprised in 2024 how close the Cosmos ecosystem is to this reality when combined with native fee abstraction that allows a Cosmos chain to accept any fee token approved by governance.

Prior to 2023, IBC was siloed to the Cosmos ecosystem. In 2023, a number of teams have been working to expand IBC proliferation into the Ethereum ecosystem and its L2s, as well as other ecosystems. The Composable Finance team was the first to export IBC outside of Cosmos by implementing a solution to Polkadot/Kusama. Since then, multiple teams like Union Protocol, Polymer, Composable, and Electron Labs are all working on their own zkIBC implementation to bring IBC to Ethereum and its L2s, as well as a team working on bringing IBC to Avalanche. 2024 could be the year of real IBC proliferation and abstraction based on momentum seen in 2023.

Final Thoughts

2023 was a hell of a year for the crypto industry. We laughed, we cried, we “four”ed, but we survived. Bitcoin’s ETF narrative, mixed with an unexpected spark in Ordinal activity, reminded speculators not to lose sight of the king. The L2 race heated up significantly with the launch of Scroll, Linea, Mantle, and many others. Celestia finally launched on mainnet and picked up the modular momentum that was pioneered earlier by the Cosmos ecosystem. Alt L1s saw their day, as sentiment shifted away from only favoring Ethereum and giving more credence to other potential designs. Regardless of whatever camp one sits in, the days of paying excessive fees to process a transaction are in the rearview. These were just a few of the major milestones throughout 2023. With institutional flows primed to enter the market and an upcoming Bitcoin halvening, 2024 is set up to be a remarkable year.

This report is sponsored by Marinade.

Marinade is a stake automation platform that monitors all Solana validators and delegates to 100+ of the best-performing ones. Founded at the 2021 Solana x Serum Hackathon, Marinade enables users to earn passive staking rewards through its high-performance delegation strategy while contributing to the security and decentralization of the network. Marinade's liquid staking token is mSOL, a collateralized and fully liquid version of SOL a user receives when they stake their SOL and can be used throughout Solana DeFi. Marinade is also the creator of Marinade Native, the first product on Solana that offers the performant and permissionless staking strategy to 100+ validators, but without smart contract risk. Users stake over 10 million SOL with Marinade, making it the #1 protocol by TVL on Solana.  Learn more here.