The Internet Capital Markets (ICM) Roadmap

This article is a summary of Anza's "The Internet Capital Markets Roadmap" https://www.anza.xyz/blog/the-internet-capital-markets-roadmap. Please refer to the original post for details.

Introduction

Solana's original mission was to build the decentralized foundation for "Internet Capital Markets (ICM)". Increasing bandwidth and reducing latency (IBRL) is essential, but not sufficient on its own. Solana's roadmap needs a third pillar: addressing "market microstructure", the fine-grained design of how markets work.

Markets involve countless interlocking elements, each of which has a large effect on the overall equilibrium. Until now, it was not clear how market structure in ICM should differ from traditional finance (TradFi).

Now, however, the Solana ecosystem is converging on a vision called "Application-Controlled Execution (ACE)". ACE's ultimate goal is to let smart contracts control transaction ordering at millisecond granularity. In conversations with teams across the ecosystem, there is consensus that market structure is Solana's biggest challenge.

The entire Solana ecosystem is working on solving it.

Today Solana is the best platform in crypto on every front: users, liquidity, regulatory clarity, excellent wallets, and a global onboarding environment. Yet some applications choose to build on other technologies in order to experiment with their own market structure. Solana must be not just a place for user access, but "the first choice for anyone who wants to experiment with new market structures".

The rest of this post is divided into two main sections:

1. An overview of the key trade-offs in market microstructure

2. Concrete short-, medium- and long-term solutions to enable flexible market structure on Solana mainnet

Let's begin.

Trade-offs in market microstructure

The following is not exhaustive, but these are the typical trade-offs apps face in the industry today.

Privacy vs. transparency

Should orders be made public before execution?

• If public: asymmetry with market makers decreases and execution quality for retail orders improves.

• If private: market makers can avoid risk, but market transparency and predictability decline.

Speed bumps vs. unrestricted trading

Speed bumps (order delays) reduce toxic flow and tighten spreads, but reduce volume and slow price discovery. Fundamentally, "liquidity matters more than volume", and Solana should aim to be "the most liquid market".

Inclusion vs. finality vs. execution latency

• Inclusion time (time until a transaction lands on chain): the most important metric for market makers.

• Finality (time until final confirmation): currently about 1 second, about 150 ms after Alpenglow.

• Inclusion is planned to come down to 50–100 ms by early 2026.

Colocation vs. geographic distribution

• Colocation (the same data center) looks fast at first glance, but information about geographic events takes time to propagate.

• MCL (Multiple Concurrent Leaders) makes it possible to reflect information from around the world in the same slot simultaneously.

Maker priority vs. taker priority

• Maker priority leads to tighter spreads.

• Solana currently has no explicit priority, but in practice takers have the advantage.

• With ACE, this becomes controllable per app.

Retail vs. institutional

• Retail is welcome because it adds uninformed flow and tightens spreads.

• On Solana, both can coexist if each app provides a UX optimized for its users.

Flexibility vs. opinionated design

• Solana does not believe "one market structure is best".

• It builds a foundation for rapid cycles of real-world testing, data collection and improvement.

Hybrid vs. fully on-chain

• Solana's goal is not to be a complementary settlement layer for CEXs, but to build fully on-chain markets.

The ICM roadmap: short-, medium- and long-term outlook

Short term (the next 1 to 3 months)

Jito's BAM (Block Assembly Marketplace)

• BAM is a next-generation transaction processing layer where plugins define per-app transaction ordering.

• It runs inside a Trusted Execution Environment (TEE), combining privacy with guaranteed transparency.

• Drift, Pyth, Dflow and others have already joined as design partners.

• With BAM, CLOBs on Solana reach CEX level.

Anza's transaction landing improvements

• Agave 2.3 improves the TPU client so transactions land reliably in zero slots.

• Fixes around QUIC have greatly improved reliability compared to before.

Medium term (the next 3 to 9 months)

DoubleZero (DZ)

• A dedicated fiber network that goes beyond the public internet.

• Multicast and hardware acceleration deliver 10x bandwidth and up to 100 ms lower latency.

• Mainnet rollout planned for September, followed by gradual adoption.

Alpenglow

• Solana's new high-speed consensus protocol.

• Finality goes from the current 12.8 seconds to 150 ms.

• This makes implementing MCL and asynchronous execution easier.

APE (Asynchronous Program Execution)

• Fundamentally reduces latency by making transaction execution asynchronous.

• Planned for mainnet in the first half of 2026, after Alpenglow.

Long term (2027 onward)

MCL (Multiple Concurrent Leaders) and protocol-level ACE

• The foundation for Solana to give apps full control over ordering.

• Not depending on a single leader delivers censorship resistance and fair ordering.

• Ordering by priority fee enables logic such as cancel priority.

• Information from around the world can be reflected in the same block at once, a structure impossible for centralized systems.

Toward a chain that synchronizes the world's information

MCL is "the only way to ingest the world's information fastest and process it simultaneously". The tools Solana provides build financial market infrastructure for the internet age that centralized exchanges cannot replicate.