Wartime Arc
A lot can change fast in crypto, and the current state of things presents an opportunity where the trends and narratives that have been driving the market have been discounted in light of the $TRUMP memecoin launch. As liquidity recovers and optimism likely returns, assuming Trump includes crypto in his day-one executive orders, the market may quickly rally from this strange lowpoint that most tokens have found themselves in.

While memes have been further vindicated, AI is still the leading sector by mindshare and AI-related names potentially have the best risk/reward here, just as they did during the last market dump. One of the more interesting tokens that is most in tune with the AI agent’s narrative is $ARC. As ai16z doubled down on typescript, Arc’s rig provides a rust-based framework for developers to create their own lightweight AI agents, already establishing itself as the second-most adopted open-source agent framework.
Key Takeaways
- The market has reset a bit in light of $TRUMP, both in terms of price and mindshare, presenting an opportunity for winners to emerge.
- The AI agents narrative is only heating up, and infrastructure providers still have room for growth, though frameworks and platforms for agent development are quickly becoming commonplace. Winners will need to differentiate themselves through concentration and partnerships/integrations.
- Arc has been heads-down building, providing a Rust-based solution for building lightweight agents, now focusing on expanding integrations on the development-side as well as with EVM chains.
- $ARC Mindshare and price have been steadily increasing, showing little signs of froth relative to other projects’ adoption and price action.

State of the Market & AI Agents Landscape
Due to the effect it has had on the market, $TRUMP needs to be addressed. Onchain tokens have seen significant drawdowns in the days and hours following the launch of $TRUMP and $MELANIA. However, the temporary shock effect of a second memecoin launching has temporarily distracted market participants from the greater growing-of-the-pie that these Presidential memecoins will likely result in, particularly for on-chain tokens.
This has created an interesting situation where the total crypto marketcap is likely to significantly increase especially post-inauguration, yet most tokens are down significantly, some worse than others. Mindshare has also seemingly been somewhat reset, with everyone’s timeline being dominated by all things Trump for the past few days, past successes may matter less in how market participants choose to allocate to AI agent tokens.
Outside of factors affecting the broader crypto and specifically on-chain markets, the AI agents narrative isn’t going anywhere. The total marketcap of the sector has been hit hard, currently sitting at $14.3B, down from a peak of just under $20B. At the start of this craze in crypto, back in October and November, many might have thought that agents were a mostly crypto-native trend. But since our previous coverage, the likes of Sam Altman of OpenAI, Jensen Huang of Nvidia, and other key figures have cast extremely optimistic expectations for the role of agents in the growth of AI in the year ahead.
“We are now confident we know how to build AGI as we have traditionally understood it. We believe that, in 2025, we may see the first AI agents “join the workforce” and materially change the output of companies. We continue to believe that iteratively putting great tools in the hands of people leads to great, broadly-distributed outcomes.”
-OpenAI CEO Sam Altman

Even as the total Agents marketcap has grown, it is still under $15B. There is likely a discrepancy between the total AI agents marketcap and the amount of attention this genre as a whole has recently started receiving, with crypto being one of the few ways to express an AI agent trade in isolation.

All of this is great for AI agents as a whole, but where Arc fits in is a more specific niche, more dependent on the team’s strong performance and execution. However, the project can also see tailwinds from greater on-chain trends that align with the direction the team is building.

When analyzing the trends in AI agents’ valuations recently, we can see that Solana has now outpaced Base in terms of Agent marketcap valuations. Meanwhile, the marketcap on other chains, primarily Ethereum mainnet, has also grown to be a larger percentage of the total sector’s marketcap. Both the launch of $TRUMP on Solana, which sparked the coin to reach new ATHs, as well as some renewed leadership for Ethereum, seem likely to continue this current trajectory.
A lot of these frameworks, whether it be ai16z or Arc, are actually chain-agnostic in nature. While something like Arc might seem like a natural fit for Solana considering it is based on Rust, integrations that the team has already begun adding can allow for these agents to deploy across the EVM ecosystem as well.
Still, chains act as magnets for attention and liquidity. As seen with Virtuals, what chain a project deploys on can make all the difference when it comes to recognition and valuation, which can also drive development and funding, even if the agent itself isn’t bound to one chain.
In the current state of the market, $ARC has faced a large drawdown along with the rest of the market, though already quickly rebounding off the lows. If purely judging by the chart, the token looks primed to make new highs; the latest dip caused by the $TRUMP liquidity suck might not last too long. $ARC has actually been one of the best performers in the post $TRUMP sessions.

The current iteration of the AI agent space has more tokens than one can keep track of, making the ability to find framework or infrastructure-related tokens that have a better potential for risk-adjusted returns important. On this front, $VIRTUAL and $AI16Z have been the main leaders, with $AIXBT, an individual agent now also reaching a valuation of over $700M.
Now we’re also seeing the DeFAI category emerge, with its own category leaders including $ANON and $GRIFFAIN fetching large valuations. Beyond these tokens, there are a number of agents rolling out their own infrastructure, including $BULLY and $ZEREBRO, that have been able to fetch high valuations. Of the current AI agent frameworks, $ARC stands out as one of the leaders in terms of adoption. In fact, Arc is actually the main leader, second only to ai16z when it comes to GitHub stars rankings.

Arc’s Offerings
Rig, by Arc, is a Rust-specific framework for building LLM applications and agents. Rust is selected for its performance, safety/bug prevention, and concurrent nature. The goal here is to take the functions of top LLM models from OpenAI, Anthropic, Perplexity, Gemini, and Cohere, and channel it through custom agents via one unified API. Users have the option to implement their own separate models if they wish. Rig effectively serves as a modular stack for creating apps and agents and deploying them on-chain, tying existing LLM integrations together.
Rig provides open, core Rust libraries that developers can use for creating their own agents, as well as infrastructure for multi-agent management and swarms. Various abstractions are made accessible so that developers can spend more time building custom tools and features that are specific to what they want in their end-product, and less time on basic functionalities. Retrieval-Augmented-Generation (RAG) is leveraged to pool data from dynamic knowledge bases, allowing for agents to be more current in their interactions and adjust their outputs accordingly.

As far as tokenomics go, like its peers, $ARC was created on Pump.fun, with a marketcap and FDV of $433M, slightly down from a peak of ~$530M. The $ARC token does have some allocation for contributors via prize pools. Similar to ai16z bounties, this is used to incentivize participation and meaningful growth, likely a net positive for the ecosystem even accounting for the sell pressure.
Talk is Cheap
Playgrounds, the creators of the rig framework, and its Co-founder and CEO Tachi, are relatively quiet on X, at least much less active than the teams behind the other leading frameworks and platforms. The team has strayed away from making lofty announcements, expressing ambitious goals or otherwise introducing more ‘flashy’ initiatives when compared to the likes of Virtuals, ai16z, Zerebro, or even something like Dolos ($BULLY).
This isn’t necessarily a bad thing and it may present an accumulation opportunity if one expects agent developments and continued partnerships in the future which have more substance that the market might not initially have expected.

Arc mindshare has only consistently increased since its launch, with price now lagging mindshare. Even after hitting ATHs, $ARC price has not become over-extended relative to mindshare. Without a native agent and being a new framework that has not seen a large amount of products yet built on it that consumers can actually interact with, Arc naturally has less attention coming its way. However, this is changing as of late, with engagement up nearly 60% in the past week, impressions also seeing a large bump.

Source: Cookie.fun – Increasing $ARC engagement and impressions; the project still somewhat lacks an X presence that similarly sized projects have, despite its growing numbers.
The team is also looking to become more of a hub with Arc Complex, an initiative perhaps inspired by ai16z partners and that broader community. This is simply a means for the Arc team to build a developer ecosystem and promote collaboration between individuals and teams, as well as shine a spotlight on what some people are building, which will reflect well on the underlying rig framework. The intention for involving agents in the actual collaboration and development of projects has also been expressed.
Users can apply to join the Arc Complex registry, potentially receiving funding in $ARC or $SOL to create their own LP pair for their token. In its current stages, this initiative is obviously very rudimentary compared to something like Virtuals, but this could be a vector of growth to look out for. The team is also explicitly scouting out projects as well to try and add to the Arc Complex.
There might be something to be said about exclusivity; project submissions and applications are actively curated, and not just anyone can become a part of this collective. This incubation aspect of Arc Complex will have more teams launch projects in the near future, providing more examples of the rig framework in action and drawing attention to the framework as a whole.
Ecosystem Building
Ecosystem integrations are what allow Arc to differentiate itself, apart from its established developer adoption, and defend its spot against new platforms and frameworks. These ‘brand x brand’ partnerships might seem somewhat superficial, but they also bring attention to the rig framework itself, and in Arc’s case, provide real utility for developers. These integrations will enhance the rig framework’s capabilities, increase distribution, and allow for deployment on more chains.

Arc announced its integrations and ecosystem partners more or less all at once; the market had some time to digest this, though this was cut a bit short by the launch of $TRUMP. Considering the team’s perceived reluctance to drop hints ahead of time, repricings can potentially happen all at once, perhaps making $ARC a better hold or swing trade and not something to go in and out of. Some of these integrations are common, and will include other peer frameworks, while others are more exclusive; both are good for Arc’s growth, considering it flew under the radar compared to more understandable agents, until now.
Timing and Invalidations
Now is a good time to scale into high-conviction bets, considering that almost all tokens are down across the board. $ARC’s drawdown isn’t as large as other tokens, but still significantly off from its highs.
When it comes to timing, it’s also worth noting that $ARC price action has been more slow-and-steady compared to other leaders, displaying less exaggerated drawdowns for now. As far as execution risk, the team is less keen on being in headlines as opposed to something like ai16z, and a case can be made that $ARC is a better hold instead of something to constantly trade lows and highs.
That being said, token price should likely be dependent on continued growth in adoption, as well as any new announcements and partnerships that come out in the near future. It will be prudent to monitor these and consider them as main drivers of price.
Like with most tokens we cover, time-based invalidation is another aspect to consider. While it might not look likely right now, even tokens at the top of the AI mindshare charts have the potential to stagnate, whether it’s to accommodate new AI agent plays with more momentum or if the sector itself stagnates. The former is more likely than the latter.
Like any other AI agent token, the market can turn on a dime and leave tokens in the past, so it’s important that Arc maintains its adoption, or relevancy in the absence of growth in adoption, during pivotal points where the AI agent marketcap exponentially increases. A ceiling for something like Arc is claiming the spot as a clear third in the agents infrastructure space, below $AI16Z and $VIRTUAL. At this point, there would likely be little perceived upside, with $ARC having surpassed tokens like $FAI and $GRIFAIN, as well as individual agents like $AIXBT, which have more of an X presence and get more attention. These relative rankings may be a better gauge of performance, given that the AI agents total marketcap has moved extremely unpredictably lately.

Conclusion
The AI agents space is likely to be the sector with the most sustainable momentum in 2025, with the launch of $TRUMP being a blip in this trend that also serves to bring more attention to Solana and on-chain tokens compared to others. With mindshare, engagement, partnerships, and price action all looking strong, $ARC appears to be one of the best bets in the current market, serving a clear niche: rust-based, lightweight AI agents.
As more AI agent frameworks come to market, these platforms for developers to build on will have value diluted, with concentration and partnerships being two primary vectors through which infrastructure can set itself apart. This is more or less also what we’ve seen play out between competing L1s and L2s.
In the past week or so, the case for $ARC has only strengthened with numerous partnerships being inked. Builders in the Agentic AI space have often been quick to partner with each other and combine efforts, and now $ARC has begun partnering with external chains and service providers. This shows a new side to the Arc team, that they are indeed BD-focused and will take it upon themselves to actively bring their service to developers. It’s still early days, but the Arc team has also demonstrated their focus on creating a collective, a community that developers perceived as valuable, even scouting existing projects to join the Arc Complex.
Outside of ai16z, ARC is already the leading AI agent framework in terms of adoption, with ongoing BD and partnerships, community building, and integrations on track to maintain or extend this lead.
Disclosures
Alea Research has never had a commercial relationship with AI Rig Complex and this report was not paid for or commissioned in any way. Members of the Alea Research team, including those directly involved in the analysis above, may have positions in the tokens discussed. This content is provided for educational purposes only and does not constitute financial or investment advice. You should do your own research and only invest what you can afford to lose. Alea Research is a research platform and not an investment or financial advisor.


