Coffer

Cube TVL just doubled 🎉

@vlzhr also launched a new 3× leveraged multi-token pool SOL/USDC/CBBTC.

The more pools and liquidity inside Cube, the more swap routes become available and the better execution traders get.

This is only the beginning. Soon, we’ll fully open pool creation for all users.

So if you want to try Cube, you can already start supplying liquidity to existing pools and earn XP.

App | X | Telegram | Chat | Docs

There will never be a better time to farm XP 👀

Cube XP is live. Rewards program for LPs and traders. Epoch 1 just started.

For LP providers:
• $1 of liquidity = 50 XP per day, every day it stays in the pool

For swappers:
• $1 of LP fees from swaps = 10,000 XP

Pools with higher fees produce more XP per swap, helping keep rewards fair and balanced.

XP runs in 4-month epochs. After each one, rates halve. Which means Epoch 1 right now is the most rewarding it'll ever be.

Halving countdown and leaderboard are both live in the app. Lock in your rank early.

Long term, the leaderboard is just the start. We're building a full rewards system for everyone who uses Cube. If Cube ever introduces a token, XP may eventually be convertible into it.

Epoch 1 won't last forever. Get in early.
Docs → docs.cubee.ee/rewards/cube-xp
Leaderboard → cubee.ee/leaderboard

App | X | Telegram | Chat | Docs

50/50 pools were revolutionary in 2018, but not anymore...

The idea was simple: x*y = k

Where:
• x = amount of token A in the pool
• y = amount of token B
• k = a constant the pool tries to maintain

As one asset gets bought, the other automatically becomes more expensive.

No order books. No market makers. Just programmable onchain liquidity.

For the first time, anyone could become a liquidity provider. And it changed DeFi forever.

But the market evolved, and today we no longer trade just simple pairs. We now have:
- BTC wrappers
- LSTs
- synthetic assets
- dozens of correlated assets

And this is where traditional 50/50 pools start showing their limitations.

Most AMMs still treat assets as if they were identical. But modern assets have:
- different redemption mechanics
- different risk profiles
- different liquidity conditions
- persistent premiums and discounts

The result is persistent pool imbalances, lower capital efficiency for LPs, and worse execution for users.

The next generation of AMMs won’t just optimize trading anymore. It will optimize liquidity itself.

At Cube, we’re building multi-token pools with virtual balances and per-asset configurations, allowing pools to account for persistent discounts/premiums instead of forcing everything into a naive 1:1 model.

App | X | Telegram | Chat | Docs

We’ve been on mainnet for two weeks, and someone already tried to take us down!

This week we registered a DDoS attack – 6 million requests in a short period of time. For an early-stage protocol, that’s serious load. The team responded fast, the infrastructure held, 15 minutes from attack to fix.

Behind Cube is a technical team with a strong background. People who built high-load systems long before this project existed.
We’re just getting started, TVL is growing, first pools are live.

We’ve already been through a baptism by fire and keep building 😉​​​​​​​​​​​​​​​​

App | X | Telegram | Chat | Docs

Remember our post about BTC wrapper pools and the imbalance problem? We’ve deployed the pool.

This isn’t just another multi-token pool. This configuration doesn’t exist anywhere – not on Solana, not on any other chain.

Here’s what’s inside:
The pool works like Balancer with concentrated liquidity, but tokens are pegged not 1:1 – but 1:0.998. Unlike Curve, which always pushes assets back to 1:1 parity, this pool holds a target ratio of 1:0.998 with amplification of ~2000. Fundamentally different result.

This matters for BTC wrappers that trade at a persistent discount due to redemption fees. The pool reflects real market pricing – not forced parity.
Token weights are configured individually based on the liquidity and activity of each asset. This level of precision doesn’t exist in any other protocol or on any other chain.

We seeded $10,000 in liquidity. LPs are welcome to join.

App | X | Telegram | Chat | Docs

Part 2

5) DIRECT COMPARISON

Same 0.1 LBTC → BTC swap, fair value 0.102 BTC:

Pool /// Out (BTC) /// vs Fair
──────────────
Balancer 50/50 (10/10) /// 0.09900990 /// -2.94%
Balancer 50/50 (10/10.2) /// 0.10099010 /// -1.00%
Curve A=100 /// 0.10196979 /// -0.03%
Curve A=1000 /// 0.10199695 /// -0.001%
Cube (vLBTC=10k / vBTC=10.2k) /// 0.10199898 /// -0.001%

Cube matches Curve at A=1000 — without needing a StableSwap invariant. It's a generalized weighted AMM with per-token risk tuning.

—

6) WHY THIS MATTERS BEYOND ONE SWAP

In a Cube BTC pool with up to 10 wrappers, each token can have its own leverage matching its real-market price:

- cbBTC (free 1:1 redemption) → leverage 1020
- WBTC (20 bps redemption fee) → leverage 1018
- LBTC (bridge / withdrawal risk) → leverage 1000
- new wrapper (untrusted, high risk) → leverage 600

The pool encodes risk-adjusted fair value per asset. Arbitrage opportunities collapse. LPs farm organic volatility, not toxic flow.

And because it's one pool with 10 assets instead of 45
isolated pairs, routing is dramatically simpler. Arbitrage
volume between wrappers stays inside the pool as LP fees instead of leaking to MEV bots.

—

7) THE BIGGER PICTURE — cubee.ee

The BTC wrapper case is one application of a more general design. Cube combines three primitives from the past 5 years of DEX evolution:

- Weighted pools from Balancer (custom token ratios)
- Concentrated liquidity from Uniswap V3 (capital efficiency)
- Generalized invariant control (per-token leverage)

Multi-token pools at Cube reuse the same liquidity across many pairs.

One LP position in a SOL/USDC/BTC/BONK pool serves swap routes for every combination — SOL/USDC, SOL/BTC, BTC/BONK, USDC/BONK. Your capital keeps earning even when individual pairs go idle.

The V3 inactive-bin problem disappears. In standard concentrated liquidity, >99% of deposited capital often sits in inactive bins earning zero. Cube pools route the same liquidity through multiple pairs simultaneously, so idle capital becomes active capital somewhere else in the pool.

For traders: smaller slippage, smart order routing across all internal pairs, deeper effective liquidity than the TVL suggests.

For LPs: leveraged fees ($100 in a 10x pool earns fees like $1,000 in a classic AMM), multi-asset exposure from a single position, no idle capital.

—

8) WHAT'S NEXT

A dedicated BTC pool is coming soon, with per-wrapper leverage configurations designed around the math above.

If you've hit real on-chain problems that broke other AMM designs — DM us. We read everything.

Live on mainnet: cubee.ee

App | X | Telegram | Chat | Docs

Part 1

An LP with 2+ years of experience LPing BTC wrappers reached
out to us last week with a problem.

We sat down, did the math, and realized Cube's pool design
solves it almost perfectly.

👇 Here's the breakdown.

—

1) CONTEXT

Cube is live on Solana mainnet. Security review by Serokell completed before launch. Real pools, real liquidity, real swaps happening right now.

This isn't a thesis post. It's a walkthrough of how we already fixed a problem that broke standard pool designs.

—

2) THE PROBLEM

cbBTC, WBTC, xBTC, LBTC — they all represent BTC. But they don't trade equally.

Each wrapper has its own:
- redemption mechanics
- withdrawal delays
- issuer / custody risk
- on-chain liquidity depth

This creates persistent price deviations. A wrapper might consistently trade at 0.995 BTC on the market while standard multi-token pools still price it at 1.000.

The result: free arbitrage. Buy wrapper at 0.995 → deposit into pool → withdraw clean BTC at 1.000. Repeat.

LPs end up on the losing side every time. The pool fills with the discounted asset. Toxic flow extraction.

—

3) WHY EXISTING DESIGNS DON'T FIX THIS

A standard Balancer 50/50 pool with 10 LBTC / 10 BTC will quote a starting spot price of 1.02 (correct).

But for a 0.1 LBTC → BTC swap:
out = 10 * 0.1 / (10 + 0.1) = 0.0990099 BTC

Fair value: 0.1 × 1.02 = 0.102 BTC
Slippage: ~2.93% worse than fair

The pool prices the trade as if both assets were 1:1, ignoring the actual peg. Balancer's weighted formula can't encode the real exchange rate without an external rate oracle.

Curve solves this with stable-swap amplification (A).
At A=1000–5000, Curve gets very close to fair price (~0.10199 BTC for the same swap). But Curve's invariant is rigid — one global A for the entire pool, no per-token control.

—

4) HOW CUBE SOLVES IT

Each token in a Cube pool has TWO configurable parameters:
- individual weight
- virtual balance (leverage)

The swap formula uses virtual balances instead of real ones:
out = vOut * (1 - (vIn / (vIn + amountIn)) ^ (wIn / wOut))

Set per-token leverage so virtual balance ratios encode the fair price directly:

Real: 10 LBTC + 10 BTC
Leverage: 1000 (LBTC) / 1020 (BTC)
Virtual: 10,000 vLBTC / 10,200 vBTC

→ Virtual price = 10200/10000 = 1.02 (fair)
→ Depth: feels like a 10,000 LBTC pool

For 0.1 LBTC → BTC:
out = 10200 * 0.1 / 10000.1 = 0.10199898 BTC
Slippage: ~0.001%

The first Cube pool is now live

Our first multi-token pool includes:
• $JITOSOL
• $SOL
• $USDC
• $USDT

Mainnet has just launched, but we’ve already added the first liquidity – and you can start swapping tokens on Cube right now ;)

More pools, updates, and new features are coming soon

Support Cube and try out the first pool

App | X | Telegram | Chat | Docs

Mainnet is live 🎉

We’ve the first stage of development of Cubee and passed a deep security review with Serokell ahead of launch

Cubee is now live on Solana mainnet. More updates and improvements are coming

We’re also rolling out a fully redesigned landing page – with strong visual design and bullet points

Here’s a part of what you’ll find on the page:
• real-time key metrics
• benefits for traders & LPs
• documentation and open-source SDK

Support Cubee on mainnet and check out the updated landing 👀

App | X | Telegram | Chat | Docs

Marketing: we’re gonna fucking disrupt the space
Reality: $2 TVL 😂

We’ve developed and open-sourced the Cube SDK

It is now publicly available on GitHub and provides access to the protocol’s infrastructure for integrations

The SDK allows fetching data directly from smart contracts and the Cubee backend

The backend is now also public – anyone can access it and retrieve the required data

The SDK wraps all this logic into clear classes, simplifying development and integration

This makes it easy to integrate for:
• AMM bots
• solvers
• developers in general

The SDK is now available 👉 https://github.com/cubee-ee/sdk

App | X | Telegram | Chat | Docs

We’ve just completed a deep code & architecture review with Serokell ahead of mainnet.

Cube – concentrated liquidity DEX on Solana that combines multi-token pools and leverage, natively giving high double-digit APYs (achieved in real tests).

Scope included:
• core smart contracts
• execution logic (incl. virtual balances / leverage mechanics)
• edge cases, attack surfaces, performance

Serokell is an engineering and research team with 11 years of experience in blockchain security. They bring experience from projects like 1inch and other complex DeFi systems to our dex!

Cube is now in a much stronger position to ship.

App | X | Telegram | Chat | Docs

😁

Based on a study of 11 different liquidity pools (ranging from 2 to 10 tokens, with varying compositions and lifecycle durations), the following conclusions were reached:

• In 2-token pools, price impact is significantly more pronounced compared to multi-token pools
• Growth of low cap tokens can be considered an advantage (positive slippage)
• Tokens with low trading volume and insufficient liquidity reduce overall APY
• Small pools (< $5k) generate lower returns, because there are no major transactions going through them
• Stable pools with high leverage (x100) provide more consistent but not maximum APR

Conclusion:
A higher number of tokens in a pool expands the space of strategies and potential yield scenarios
However, without proper analysis of pool composition and parameters, it may lead to reduced efficiency. Therefore, careful token selection and range configuration aligned with market conditions is required

Source: https://medium.com/@izhur27/inventing-ranges-pt-4-quantitative-research-dc7fc95c0241

App | X | Telegram | Chat | Docs