Presented by
The Energy Network uses household devices to increase grid capacity. Launched by Fuse, a $400M ARR scale-up. Check out the first live product demo here.

BTC Holds - Bitcoin traded around $78,400 and Ethereum near $2,290 during Asia hours, holding key price support.

Ark Invest $72m - Ark Invest, led by Cathie Wood, purchased approximately $72 million in crypto-related stocks across its ARKF, ARKK, and ARKW funds amidst renewed Bitcoin market volatility.

Hyperliquid in Prediction Market - Decentralized exchange Hyperliquid is launching HIP-4 markets enabling prediction-market-like outcome trading.

New here? Ask Copilot up to 5 free queries per day with Messari Basic.

How likely are you to evaluate an enhanced Messari advisory service in the next 6 months?

Login or Subscribe to participate

Messari's protocol reports give you a deep dive on the foundation and state of top crypto protocols, including key metrics and notable events. See the complete list of protocol reports here and get a preview of our latest report below.

Eric Manoukian  (@CryptoRick98)  ·  Research Analyst

  • Chromia continued to position its Layer-1 as infrastructure for data-intensive applications, with Q4 execution focused on strengthening core network participation and expanding its AI and data tooling surface.

  • Chromia broadened its AI and data infrastructure footprint while bringing AI infrastructure closer to production readiness through three milestones, completing Base and Arbitrum interoperability on Dec. 5, 2025, integrating with Virtuals’ Agent Commerce Protocol on Dec. 11, 2025 to support structured, queryable agent job data on Chromia, and launching a GPU-powered inference demo on testnet on Dec. 23, 2025.

  • Chromia’s network activity expanded in Q4, with average daily transactions increasing 26.6% QoQ and 71.7% YoY to 642,900, while average daily active addresses increased 3.4% QoQ and 106.6% YoY to 4,128.

  • Chromia initiated the scheduled phaseout of staking rewards on Ethereum and BNB Smart Chain on Dec. 2, 2025, shifting staking incentives toward Chromia mainnet and aligning long-term incentives with native staking ahead of governance activation.

  • The Chromia team shipped a major redesign of the Chromia Vault in Q4, introducing portfolio-style balance aggregation across application chains, improved application discovery, and improved staking transparency through dynamic fee sharing and estimated APR displays.

Youssef Haidar  @0xYoussef_ · Research Analyst

  • BlackRock tokenized $500 million on Avalanche through BUIDL, its onchain money market fund. This deployment increased RWA TVL 68.6% QoQ and 949.3% YoY to $1.33 billion at year's end.

  • The Avalanche ETF race accelerated as Bitwise and VanEck incorporated staking into their S-1 filings. Bitwise’s BAVA proposal notably includes plans to stake up to 70% of holdings, marking a significant shift in how regulated investment vehicles approach native network yield.

  • TIS Inc., one of Japan's largest payment firms, launched a multi-token platform on Avalanche via AvaCloud. The platform enables Japanese banks to issue and manage stablecoins, tokenized deposits, and digital securities.

  • The Granite upgrade officially activated on Nov. 19, 2025, introducing dynamic block times and native biometric authentication. These technical milestones allow for sub-two-second transaction settlement and let users sign transactions via FaceID or TouchID, drastically lowering the friction for mainstream adoption.

  • Avalanche’s DeFi ecosystem showed resilience as TVL denominated in AVAX rose 41.9% QoQ to 102.8 million AVAX. This indicates a "sticky" liquidity base, as users maintained stablecoin and non-native positions in DeFi protocols on Avalanche instead of bridging value elsewhere.

Dillon Shirley @zcb_spec ·  Research Analyst

  • FLock shipped the AI Arena v2.1 update, which tightened incentive alignment by removing delegation-driven reward dilution for operators and smoothing the unlock schedule to minimize volatility.

  • The number of training and validator nodes is up 2871.4% YoY and 1523.5%, respectively. The FLock network now has 208 training nodes and 276 validator nodes supporting the network

  • FLock assisted 7 finalists in the second cohort of the United Nations Development Programme’s Sustainable Development Goals Blockchain Accelerator in Q4 2025.

  • Training and validator nodes received $2.7 million in FLOCK for participating in the AI Arena.

  • FLock.io is integrating the x402 protocol to enable agentic commerce with onchain payments, micropayments, and autonomous economic capabilities for its decentralized AI agents

From Our Sponsor

By: Dylan Bayne (@dylangbane)

An estimated $5.6T will be required to finance AI capital expenditures by 2030. In 2026 alone, hyperscalers are expected to spend roughly $602B on AI related capex, up 45% from $415B in 2025.

As capex spending accelerates, debt financing has played a surprisingly small role. While CoreWeave and other neoclouds have secured ten figure private credit facilities for compute, credit still represents a limited share of total AI CapEx to date.

The core constraint is GPU depreciation. Large scale demand for GPUs only emerged in the past three years following the launch of ChatGPT, leaving lenders with little historical data to underwrite these assets. Depreciation varies significantly based on utilization and workload, while rapid improvements in chip performance risk impairing the value of existing inventory. These uncertainties compound for lenders attempting to model downside risk.

Traditional banks have largely stayed on the sidelines. Regulatory constraints and slow underwriting processes have made it difficult for banks to move quickly enough to finance GPUs, leaving most lending activity to private credit funds. While private credit has supported the largest neoclouds, smaller operators seeking loans below $20M have few viable options. Because the cost of debt tends to be lower than the cost of equity, this financing gap is a meaningful constraint for smaller players.

This has given rise to InfraFi. Pioneered by USD.ai, InfraFi aims to connect onchain stablecoin liquidity with borrowers seeking financing for GPUs, typically to rent capacity to startups running training and inference workloads. The core thesis is that onchain systems can coordinate capital faster and more efficiently than traditional finance, making crypto a better fit for underwriting novel assets with limited historical data. USD.ai has reached a peak TVL of over $686M and plans to launch its governance token, CHIP, in mid February. More recently, several DePIN projects have announced InfraFi initiatives, including Daylight with DayFi for energy financing and DAWN for fixed wireless infrastructure.

That said, the fundamental challenge of GPU depreciation remains unresolved. Current approaches focus on risk mitigation rather than full pricing clarity. USD.ai’s CALIBER tokenization framework and its Queue Extractable Value (QEV) mechanism are designed to reduce redemption risk, while loan terms require a three year repayment period. USD.ai argues this is conservative, citing hyperscalers such as AWS and GCP that estimate useful lives of five to six years. Others disagree. Investors such as Michael Burry have suggested effective lifetimes closer to two years, which would make even three year loans aggressive.

Over time, clearer pricing signals may emerge from onchain GPU spot and derivatives markets. Projects such as Ornn and Squaretower Research are aggregating GPU pricing data across markets with the goal of establishing a reliable reference price. If successful, this could enable hedging via GPU derivatives and improve lender confidence. While still early, progress in pricing, verification, and market structure would materially increase the odds that InfraFi has legs.

Keep Reading