Investor presentation
Logotype for MARA Holdings Inc

MARA (MARA) Investor presentation summary

Event summary combining transcript, slides, and related documents.

Logotype for MARA Holdings Inc

Investor presentation summary

7 Apr, 2026

Strategic evolution and platform expansion

  • Transitioned from Bitcoin mining to AI infrastructure, emphasizing energy ownership and digital infrastructure optionality.

  • Developed a modular, high-density data center platform for both Bitcoin and AI/HPC workloads.

  • Formed a joint venture with Starwood to convert power-rich sites into AI, enterprise, and high-performance computing infrastructure, targeting over 1 GW of near-term IT capacity and a pathway to 2.5 GW.

  • Expanded internationally through Exaion, enabling sovereign-grade, compliant AI and HPC deployments, especially in regulated European markets.

  • Focused on energy-backed compute, leveraging predictable access to energy and modular expansion to meet rising AI and data center demand.

Financial and operational performance

  • Q4 2025 revenues decreased 6% YoY to $202M, with net loss of $1.7B compared to $528M net income in Q4 2024.

  • Adjusted EBITDA dropped to ($1.5B) from $796M YoY.

  • BTC holdings increased 20% YoY to 53,822, while BTC production fell 19% to 2,011 BTC.

  • Cost per petahash/day improved 4% YoY to $30.5, and by 36% over 11 quarters.

  • Liquidity, including cash and BTC, stood at ~$5.3B at year-end 2025.

AI infrastructure and market positioning

  • Positioned to capture value across the AI stack, from energy and infrastructure to applications, through partnerships with Starwood and Exaion.

  • AI platform integrates hyperscale scaling, sovereign enterprise, and capital/network alignment for flexible, high-return deployments.

  • Exaion provides access to regulated, compliance-driven markets, supporting stable, contracted revenue streams and GDPR-compliant operations.

  • Joint venture structure allows up to 50% investment in projects, leveraging Starwood’s capital and operational expertise for efficient, scalable growth.

  • Early customer demand for AI and HPC capacity is strong, with design, permitting, and leasing underway for over 1 GW of IT capacity.

Partial view of Summaries dataset, powered by Quartr API
AI can get things wrong. Verify important information.
All investor relations material. One API.
Learn more