this article on @katana's updated tokenomics is the first to embed an interactive chart created by @DefiLlama's LlamaAI.
here, you can adjust sliders for swap fees, perp fees, bridge fees, and more, to measure what the bottom line means for KAT buybacks.
so many ideas.
After buying back its own ecosystem, @katana is routing that ecosystem’s fees into $KAT buybacks for avKAT stakers.
At last week’s activity, that comes to about $700K a year, or 7.3% of market cap, and sequencer fees, 22% of revenue, would fund about 60% of it.
And if swap activity returns to late 2025 levels?
About $3.35M a year, roughly 1/3 of KAT's market cap.
Our follow-up spotlight looks at what the new tokenomics mean for Katana and how the design compares with buybacks across DeFi: defillama.com/research/spotl…
10k llamas strong 🦙
Our Research Notes newsletter now lands in 10,000 inboxes, covering everything DeFi through the lens of @DefiLlama data.
If you’re not one of them yet, the next edition is a click away.
defillama.com/newsletters
i've had the pleasure of getting to know the @noon_capital team over the past few years.
they took a unique route, blending private capital-generated & onchain yield with utmost transparency & open, community-wide discussion, all the while maintaining a laser focus on what they wanted to achieve.
happy to see these guys now Verified by DefiLlama.
Today, we’re launching a Verified by DefiLlama dashboard for @noon_capital.
This is our first dashboard verifying reserves that sit largely offchain. We’re applying the same rigor, standards, and transparency to those reserves that we apply to onchain TVL.
How it works:
• Onchain: every wallet and DeFi position holding Noon’s reserves, tracked in real time
• Offchain: positions verified against weekly private credit reports from an independent third party
As the onchain and offchain worlds of finance are increasingly intertwined, we're proud to work with partners to extend the transparency of DeFi to new assets.
lots being done across the crypto insurance scene, but who wants stuffy third-party underwriters with their briefcase full of caveats?
onchain capital should be covered by onchain capital, and for @Firelightfi, that means using FXRP staked by users for fees. this is what the process looks like:
→ exploit confirmed: once an event is flagged, Firelight reports the incident onchain. this freezes the exposed Cover Tokens
→ exploit report: ZeroShadow, Firelight's designated security partner, publishes an onchain exploit report within 48 hours
→ assessment: Firelight creates a loss schedule proposal, which then moves to a five-member Risk Consortium to validate. this assessment phase targets a 96-hour turnaround
→ liquidation: a protocol-owned reserved (first-loss buffer) takes the first tranche of the loss; staked vaults are only slashed if the buffer runs out
→ payout: proceeds go to the beneficiary address of each Cover Token, pro rata
what is potentially covered here? smart contract exploits (logic errors, reentrancy, access control failures, upgrade flaws), oracle failures, governance exploits, bad debt and depegs caused by mechanism failures, and redemption failures
what isn't? exchange hot-wallet thefts and wallet compromises
Customer-facing platforms spent the past year selling DeFi yield through Earn products, and curated vault TVL grew 26% to ~$10B while lending fell 39%.
One problem persists: when a protocol or vault gets hacked, depositors absorb the loss.
@Firelightfi, incubated by @SentoraHQ, lets XRP holders stake on @FlareNetworks to back cover for vaults, paying out if a protocol a vault deposits into is exploited. It now covers vaults at Kraken, Lido, and Sentora.
Our new report on why Total Value Covered belongs next to TVL: defillama.com/research/repor…
I've wanted this feature for so long. Previously the only way to get these growth rates was through the API or a custom dashboard.
The DefiLlama humepage is now, unequivocally, the best place to find DeFi alpha.
Which protocols have the fastest growing holders revenue?
Introducing the new "Growth" view on the DefiLlama homepage.
Sort by Fees, Revenue, Holders Revenue, TVL, or Price growth. Filter by a minimum. View by Protocol or by Category.
remember that post-FTX wave of 'audits' in Dec '22?
back then, firms like Mazars couldn't even provide cryptographic proof of ownership. they ended up pulling reports and exiting stage left.
Verified by @DefiLlama checks specific claims and states what it did and did not verify
In May, a dashboard offered “continuous cryptographic proof” that NUSD reserves matched liabilities.
Less than three months later, Neutrl froze redemptions. When holders could cash out in September, the price was 51 cents.
Nothing public shows the May proof was false.
From post-FTX accounting reports to live dashboards, reserve proofs tended to be accurate about something narrow and were read as saying something broad.
Verified by @DefiLlama, starting with @fraxfinance’s frxUSD, lists onchain reserves for anyone to check and verifies offchain balances at the source through read-only access:
defillama.com/research/spotl…
Over the past 30 days, RWA perps traded $914B. Where is the open interest sitting now?
Total open interest today: $20B
30d volume: $914B
CEX/DEX share of OI: 70.3%/29.7%
OI by asset group:
> Equities: $9.9B ($416.4B 30d vol)
> Precious metals: $4.9B ($198.8B 30d vol)
> Oil: $1.5B ($73B 30d vol)
> Equity indices: $1.1B ($56.3B 30d vol)
> AI/Compute infra: $1.1B ($132.4B 30d vol)
> ETFs: $746.3M ($55.8B 30d vol)
> Fiat: $363M ($20.2B 30d vol)
Public equities make up about half of OI, and precious metals about a quarter.
The most heavily traded groups relative to OI are AI & compute infrastructure (about 119x 30d volume to OI) and equity ETFs (about 70x).
Where?
> @binance: $5.2B OI ($444.7B 30d vol)
> @tradexyz: $4.12B OI ($77.3B 30d vol)
> @Gate: $1.96B OI ($40.8B 30d vol)
> @variational_io: $834M OI ($45.4B 30d vol)
> @bitget: $792M OI ($57.4B 30d vol)
Source: @DefiLlamadefillama.com/rwa/perps
Some thoughts on market makers, transparency & the @Forgd_ leaderboard:
Yesterday, a ranking of crypto market makers made the rounds on X and generated a lot of debate.
For context, the ranking was copy + pasted directly from a leaderboard built by @Forgd_, hosted on @DefiLlama & @Tokenomist_ai.
Importantly, this leaderboard benchmarks market makers engaged by blockchain projects to support liquidity for their native tokens – not prop trading firms deploying their own balance sheets.
It grades MMs based on a combination of:
→ Historical trade activity
→ KPI adherence
→ Venue coverage & treasury management capabilities
→ Size of client base
It’s also only one of several leaderboards we publish. You can separately see which desks historically provide the deepest liquidity, generate the most maker volume, maintain the tightest spreads, and demonstrate the strongest uptime. defillama.com/market-makers
Is the methodology perfect? Absolutely not.
But there are far too many people in crypto claiming to have alpha on “who the best market makers are” based almost entirely on gut instinct and anecdotal evidence.
After spending more than a decade in crypto (including time working at CEXs & market makers myself) I’ve probably been asked “Who are the best MMs?” hundreds of times.
I wanted to take a shot at answering that question with actual data.
So, some context on why we built these tools, how the data works, what the leaderboard does and does NOT tell you, and where I think we can improve 👇
1. Why we built it
One thing worth acknowledging: Forgd wasn’t the first company to recognize this problem.
Before we launched, other firms were already helping projects connect with market makers, structure deals, and better understand how their liquidity providers were performing. I think they deserve credit for helping validate an important thesis: market making was an enormous black box for most token issuers.
Where we wanted to take a different approach was accessibility.
We didn’t think projects should need to pay large advisory fees or recurring SaaS subscriptions simply to understand whether their market maker was doing its job.
So we built self-service tools that projects can use completely free of charge to:
→ Monitor market maker performance
→ Run competitive MM RFQs
→ Compare proposals
→ Benchmark historical MM performance
→ Make more informed decisions
Our belief was pretty simple: GTM infrastructure should be broadly accessible.
Today, hundreds of projects across virtually every sector and stage use Forgd’s infrastructure.
Some projects like @zama, @MeteoraAG and @metaplex have taken that a step further by making their market maker performance & engagement structures publicly transparent through products like “Universal Token Ratings”.
We think that’s a good thing. The more transparency projects are willing to provide around how their market makers trade, where they provide liquidity, and how engagements are structured, the better informed the entire industry becomes.
And because the tools are free, adoption has given us something extremely valuable: sample size.
Since launching in Q4 2024, we’ve tracked nearly 600 market maker engagements across 30+ market makers, creating what we believe is one of the largest structured datasets for understanding how crypto market makers actually perform across real engagements.
2. How does the methodology actually work?
TL;DR:
i. Projects use Forgd to monitor their market makers. For every integrated engagement, the market maker transmits trading activity to Forgd through a standardized API integration.
ii. Through these integrations, we capture things like maker fill volume, liquidity depth, spreads, KPI adherence and other engagement-specific data on a per-project, per-exchange and per-trading-pair basis.
iii. Separately, Forgd independently consumes market-wide data directly from exchanges and on-chain infrastructure. That means we can observe not only what the market maker says it did, but what was happening across the broader market.
put simply...
→ The API tells us what the market maker did.
→ Our independent market infrastructure tells us what the broader market looked like.
We can then calculate things like the MM’s share of total maker volume or total order book depth and evaluate its activity in the context of the entire market rather than looking at an isolated number.
We aggregate those observations at the project level and then across each market maker’s engagements to understand how that desk performs on average for the projects we monitor.
That allows us to compare things like:
→ Average depth of liquidity
→ Average maker / taker fill volume
→ Average spreads
→ KPI adherence / uptime
→ Engagement structures
→ Average loan sizes
→ Loan + option vs. retainer + working capital usage
→ Other commercial considerations
If you want to nerd out on the actual methodology, including examples of how raw data is transformed into project-level and MM-level observations, we published the full documentation here: forgd.com/market-maker-leade…
Some people will like the methodology. Some will hate it. That’s totally fine!
I’d much rather publish the methodology, expose it to criticism, and continuously improve it than pretend we’ve somehow solved market maker benchmarking forever!
3. There are also important limitations.
Our dataset is large, but it is not comprehensive.
Forgd has tracked nearly 600 market maker engagements since launching, but we obviously don’t capture every token project or every market maker in crypto.
Our software is free to use by design. We wanted to democratize access to tools that let projects monitor their market makers, understand how they’re trading, and make better-informed decisions without putting that infrastructure behind a paywall.
That model has given us a very robust sample size, but we can’t force every project to use it.
The same applies to market makers. Desks like @wintermute_t, @CumberlandSays, @jump_ , and @FalconXGlobal aren’t currently included because they aren’t integrated with our platform.
Does that mean they’re bad market makers? Or that they’re “hiding something”?
Absolutely not.
My experiences with these desks throughout my career has been extremely positive, and I think they can be fantastic partners for projects launching tokens.
Integration takes engineering resources, and these firms have their own development priorities. Hopefully we can convince more of them to integrate over time and make the dataset even stronger.
On the flip side, I want to give credit to desks like @flowdesk_co, @GSR_io, @ambergroup_io, @SeliniCapital, @keyrock, @jpegtrading, @tophash and others that have invested time integrating with Forgd and working with our team to improve the infrastructure.
Longer term, I think transparency around token liquidity is only going to become more important.
I’d love to see a world where projects proactively disclose who their market makers are, how those engagements are structured, and (retroactively, T + 1) how those desks actually traded across different venues.
That kind of transparency would give the investing public much better visibility into a token’s market structure, including the ability to distinguish organic liquidity and volume from liquidity and volume being subsidized through formal market maker engagements.
We’re obviously not there yet.
But the more projects & MMs that participate in initiatives like Universal Token Ratings (defillama.com/universal-toke…) & Token Transparency Framework by @Blockworks / @JasonYanowitz, the more complete the picture becomes. Our goal is to build the free infrastructure that makes providing that transparency as easy as possible.
4. A note on competition
I’ve seen plenty of criticism and debate around the methodology. That’s okay. In fact, I think it’s healthy.
We’re always looking for ways to improve the methodology, and thoughtful criticism helps us do that.
More broadly, competition is one of the healthiest forces in any industry. It forces everyone to build better products, serve customers better and continuously raise the standard. Web3 market infrastructure should be no different.
In fact, some of the companies we now compete with helped validate the original problem we set out to solve. They demonstrated that projects wanted better ways to select market makers, structure engagements and understand performance.
We simply had a different thesis around how those products should be delivered. Most notably, that critical infrastructure like MM monitoring and RFQs shouldn’t be locked behind SaaS paywalls or expensive advisory engagements.
And I don’t think any of this needs to be zero-sum.
There is plenty of room for multiple companies to build great products in the same category.
Personally, I’d rather spend our time shipping better software than publicly tearing down other builders.
If another platform develops a better methodology, feature, dataset or product, that should motivate us to improve ours. That’s how competition is supposed to work.
And we’re building well beyond market maker analytics. MM monitoring and RFQs are only one part of Forgd.
We’ve expanded the platform into tokenomics and protocol design, exchange listing intelligence, token unlock tracking, market structure analytics and novel institutional-grade research.
These problems are interconnected, and we ultimately think better data leads to better decisions throughout a token’s lifecycle.
5. I’ll end with a question.
What actually makes someone the “best” market maker?
- Is it the desk that consistently provides the most depth?
- The most maker volume?
- The tightest spreads?
- The strongest contractual KPI adherence?
- The best treasury management capabilities?
- The broadest exchange coverage?
- The projects with the best post-TGE price performance?
- Something else entirely?
@mdudas and anyone else who spends time thinking about market structure — I’d genuinely love to hear how you’d approach it. cc @hosseeb@brianhuang@ahboyash
Our leaderboard will continue to evolve as the dataset grows and we get better feedback.
The goal isn’t to declare that we’ve created the definitive answer to “Who is the best market maker?” The goal is to take one of the biggest black boxes in crypto and make it a little more transparent, measurable and data-driven.
We’ll keep publishing the methodology, listening to constructive criticism, improving the product, and letting the data speak for itself.
Yes, it's true: LlamaAI was announced 2025-11-07, so it's 356 days old, about 9 days short of one year.
Shipped since launch, from DefiLlama's own posts (I only pulled one page of its timeline, so this isn't complete):
- Aug 3: DefiLlama App on iOS and Android, plus voice dictation and bonus free queries for app users
- Aug 7: pptx, docx and xlsx exports
- Sep 9: social login and notifications in the app
- Sep 30: no-code dashboard builder that generates a dashboard from a prompt
- Oct 2: backend upgrade making LlamaAI "more analytical, smarter, and faster". It also powers "Today in DeFi" on the homepage
- Oct 19: interactive charts with sliders to stress-test scenarios, visualize LP curves and filter pools
- Oct 28: Live Mode, with voice conversations, steering a request while it runs, and progress summaries
Your point about features people won't notice is fair. The backend speed and quality upgrade shows up as a better product, not a new button.
The screenshot only shows the LlamaAI chat box with "What can I help you w…" cut off, so I can't tell what it's teasing.
defillama.com/ai/chat/shared…
It's data from @DefiLlama & @Forgd_, yes, but at face value these letter grades only tell a small part of the story.
The full Maker Maker Performance index breaks down into many sections - trading KPIs, coverage + capabilities, uptime, etc - all of which help create the composite score:
defillama.com/market-makers
it ain't even halloween and @DefiLlama is already handing out full sized chocolate bars and soda pops.
pretty soon, Pro users will be able to build their own LlamaAI llama-tars.
all part of what will be its biggest upgrade yet.
Anthropic hasn’t priced its IPO, but crypto venues are actively pricing it around the clock.
Perp futures on the company's valuation were trading well before the filing. Across 20 venues verified by @DefiLlama, they hold ~$154M in open interest and traded ~$85M in the last 24 hours.
These contracts price Anthropic at roughly $1.88T to $1.96T, nearly double the $965B valuation of its last private round in May.
Bitget, Binance, QFEX, Entropy, and tradexyz account for 83% of daily volume across tracked venues. Entropy, a market built on @HyperliquidX's HIP-3 framework, leads open interest totals with $54M, followed by Binance with $52M. In total, centralized exchanges hold $85M in OI, with Hyperliquid-based markets at $62M.
Each platform quotes the contract differently:
→ Exchange perps (Binance, Bitget, MEXC, and others) quote a valuation per unit, so a price of 1,879 means $1.879T
→ OKX offers an expiry future for Euro users that rebases once a real share count is disclosed
→ Entropy prices $1 as $1B of valuation and blends its own order book with private-market marks
→ tradexyz's ANT is quoted per expected share and is built to convert into a standard equity perp after listing
How did the filing translate to trading?
@binance traded $35.9M on filing day, 3.8x its prior two-week median, yet produced no lasting price move. At 13:00 UTC, Entropy's contract ranged from $1.86T to $1.98T and closed the hour lower than it opened.
Both venues peaked on September 9, with Entropy closing at $2.28T. By October 26, the implied valuation sat 17.5% lower on Entropy and 11.6% lower on Binance.
tradexyz's per-share contract opened at $138 against a $125 reference price, fell to $109 within its first hour, and now trades near $116.
Details in the 'fine print': none of these contracts carry shares. Prices come from thin order books, each venue sets its own oracle and settlement rules, and some can even delist or settle their contracts if the IPO is canceled.
An amended S-1 with a price range and share count is the next catalyst to watch. The countdown for November 16 is on, and it will likely bring even more volume and OI in the coming weeks.
For now, you can track Anthropic's pre-IPO markets on DefiLlama: defillama.com/pre-ipo/anthro…