A $20 Billion Valuation: Kalshi and Polymarket in an Arms Race?

marsbitОпубликовано 2026-03-09Обновлено 2026-03-09

Введение

A potential "arms race" is brewing in the prediction market sector, with industry leaders Polymarket and Kalshi each reportedly in funding talks at valuations around $20 billion. This represents a near doubling of their respective $12 billion and $11 billion valuations from late 2025. As of February 2026, the global prediction market has reached a cumulative notional trading volume of $127.5 billion. Polymarket leads with $56.07 billion, followed closely by Kalshi at $44.71 billion, together commanding 79% market share. Their growth trajectories differ: Kalshi has experienced explosive user growth, with monthly active users surging from 600,000 to over 5.1 million in 2025, while Polymarket's user growth has been steadier, peaking around 700,000 monthly active users. Kalshi's trading volume skyrocketed over 1100% in 2025, largely driven by sports contracts which account for 81% of its volume. It has secured key partnerships with platforms like Robinhood, which contributed over 50% of its volume in late 2025, and media outlets like CNBC and CNN. Polymarket maintains a stronghold in political and crypto event markets. It has formed significant partnerships with X (formerly Twitter), ICE (for a strategic investment up to $2 billion), and the UFC for exclusive prediction market data. Both platforms are also official partners of the NHL, and their data is integrated into Google's search and finance products. Their divergent strategies—Kalshi's focus on USD-compliant trading and...

Recently, according to The Wall Street Journal, the two leading prediction market platforms, Polymarket and Kalshi, have been in talks with potential investors for funding rounds, each with an estimated valuation of around $20 billion. In November 2025, it was reported that Polymarket was negotiating funding at a $12 billion valuation. That December, Kalshi completed a $1 billion funding round, bringing its valuation to $11 billion.

In just a few months, the valuations have nearly doubled again.

According to public market data and industry reports, as of the end of February 2026, the cumulative nominal trading volume of the global prediction market reached $127.5 billion, with Polymarket leading at $56.07 billion and Kalshi at $44.71 billion, together accounting for 79% of the market share.

While leading in cumulative trading volume, Kalshi showed more outstanding growth in 2025, not only reversing its market share from a minority position at the beginning of the year to over 60% but also driving its monthly active users (MAU) from 600,000 to over 5.1 million, demonstrating a faster pace of scale expansion. In contrast, Polymarket, leveraging its crypto-native advantages, maintained its global event coverage and cumulative trading lead, but its user growth was relatively steady, with peak MAU hovering around 700,000. The two core metrics—trading volume and MAU—clearly outline Kalshi's explosive catch-up and Polymarket's sustained深耕 (deep cultivation), forming a duopoly in the prediction market.

Specifically regarding trading volume performance, Kalshi's growth trajectory shows a leap from a low base to a high volume.

In 2024, Kalshi's annual nominal trading volume was approximately $1.9 billion, limited by early regulatory frameworks and market awareness, primarily driven by a few events. Entering 2025, this number surged to about $23.8 billion, a year-on-year increase of over 1100%. This explosion was directly reflected in monthly and weekly records: September single-month trading reached $2.86 billion, October further climbed to $4.39 billion, and December skyrocketed to $6.58 billion. The strong momentum continued into the beginning of 2026, with January's monthly trading exceeding $10.4 billion.

Sports event contracts contributed about 81% of this trading volume.

As of March 9, Kalshi's cumulative total trading volume exceeded $48.6 billion. Currently, its open interest hovers around $500 million.

Polymarket's trading volume reflects its earlier accumulation advantage and later steady maintenance. Dune Analytics data shows that in 2024, Polymarket's monthly nominal trading volume experienced explosive growth, reaching a historical high of $4.266 billion in October 2024, bringing its cumulative trading volume to $7.6 billion.

Subsequently, although its monthly trading volume slowly declined, it remained at a high level. The turning point occurred in September 2025, when Polymarket began its pulse-like surge in trading volume.

October single-month trading broke $4.1 billion, November broke $4.3 billion.

By 2026, in January alone, its trading volume broke $7.658 billion, and February broke $7.9 billion. As of March 9, its cumulative trading volume exceeded $59.9 billion.

According to Dune data statistics, its total number of users has skyrocketed from 40,000 in 2024 to 2.31 million today.

It is worth noting that Polymarket still holds an advantage in liquidity depth for political and crypto events, with some weekly transaction counts accounting for up to 57%, but its overall market share dropped from 95% to 35-40% in mid-2025 before gradually stabilizing.

The divergence in trading volume data between the two platforms essentially stems from differences in business focus and user access models. Kalshi, as a CFTC-regulated entity, focuses on compliant trading in USD, covering over 42 US states. Kalshi gains the upper hand in sports dominance, while Polymarket maintains a lead in political and crypto domains, jointly pushing the entire market's weekly nominal trading volume to stably remain in the tens of billions of dollars.

In terms of ecosystem partnerships, the two major prediction markets are also frequently active, deeply integrating with mainstream institutions, media, and sports IPs. Both are not only opening offline pop-up shops but also expanding疯狂 (frantically) online.

Kalshi focuses on leveraging its compliance advantages, forming a strategic partnership with Tradeweb and receiving a minority equity investment, embedding real-time probability data into its institutional trading platform; Robinhood became its largest traffic source, contributing over 50% of trading volume in the second half of 2025; simultaneously, it secured exclusive media partnerships with CNBC and CNN, with data directly integrated into programs and reports.

Polymarket, on the other hand, emphasizes Wall Street data output and entertainment penetration. In June last year, Polymarket partnered with X, becoming its official prediction partner. It secured a strategic investment of up to $2 billion from ICE, which incorporates the data into financial product streams. In November, Polymarket reached a multi-year exclusive agreement with TKO Group, becoming the official prediction market partner for UFC and Zuffa Boxing, with data integrated into broadcasts and live events.

In January 2026, Polymarket entered an exclusive partnership with Dow Jones-owned media, providing prediction market data to media outlets including Barron's and The Wall Street Journal.

Interestingly, both are official NHL prediction market partners. Additionally, Google has integrated data from both into its search and finance platforms.

Overall, judging from the trading volume and MAU data over the past two years, both are jointly driving the evolution of prediction markets from niche tools into mainstream information and risk management platforms. This trend is not only reshaping the valuation logic of the crypto industry but also providing real-time signal references for traditional finance. If subsequent funding rounds materialize, coupled with热门 (hot) geopolitical events (US-Iran conflict), World Cup tournaments, and the eve of the US election, various data points for both platforms in 2026 are expected to break historical records.

Связанные с этим вопросы

QWhat are the current valuations of Polymarket and Kalshi, and how have they changed recently?

ABoth Polymarket and Kalshi are currently in talks with potential investors at valuations around $20 billion. In November 2025, Polymarket was negotiating funding at a $12 billion valuation, and in December 2025, Kalshi completed a $1 billion financing round at an $11 billion valuation. Their valuations have nearly doubled in just a few months.

QWhich platform leads in cumulative nominal trading volume, and what is the combined market share of these two companies?

AAs of the end of February 2026, Polymarket leads with a cumulative nominal trading volume of $56.07 billion, followed by Kalshi with $44.71 billion. Together, they account for 79% of the global prediction market share.

QWhat drove Kalshi's explosive growth in trading volume in 2025?

AKalshi's trading volume surged by over 1100% in 2025, driven primarily by sports event contracts, which contributed approximately 81% of its trading volume. Key monthly records included $6.58 billion in December 2025 and over $10.4 billion in January 2026.

QHow do Polymarket and Kalshi differ in their business focus and user access models?

AKalshi, as a CFTC-regulated entity, focuses on compliant USD trading and is available in over 42 U.S. states, with a strong emphasis on sports contracts. Polymarket leverages its crypto-native advantages, maintaining leadership in political and crypto events, and uses cryptocurrency for settlements.

QWhat major partnerships have Polymarket and Kalshi established to expand their influence?

AKalshi partnered with Tradeweb for institutional data integration, Robinhood for significant traffic, and CNBC and CNN for media coverage. Polymarket collaborated with X (Twitter) as its official prediction partner, ICE for a strategic investment, TKO Group for UFC and boxing data, and Dow Jones media outlets including The Wall Street Journal.

Похожее

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

OpenAI engineer Weng Jiayi's "Heuristic Learning" experiments propose a new paradigm for Agentic AI, suggesting that intelligent agents can improve not just by training neural networks, but also by autonomously writing and refining code based on environmental feedback. In the experiment, a coding agent (powered by Codex) was tasked with developing and maintaining a programmatic strategy for the Atari game Breakout. Starting from a basic prompt, the agent iteratively wrote code, ran the game, analyzed logs and video replays to identify failures, and then modified the code. Through this engineering loop of "code-run-debug-update," it evolved a pure Python heuristic strategy that achieved a perfect score of 864 in Breakout and performed competitively with deep reinforcement learning (RL) algorithms in MuJoCo control tasks like Ant and HalfCheetah. This approach, termed Heuristic Learning (HL), contrasts with Deep RL. In HL, experience is captured in readable, modifiable code, tests, logs, and configurations—a software system—rather than being encoded solely into opaque neural network weights. This offers potential advantages in explainability, auditability for safety-critical applications, easier integration of regression tests to combat catastrophic forgetting, and more efficient sample use in early learning stages, as demonstrated in broader tests on 57 Atari games. However, the blog acknowledges clear limitations. Programmatic strategies struggle with tasks requiring long-horizon planning or complex perception (e.g., Montezuma's Revenge), areas where neural networks excel. The future vision is a hybrid architecture: specialized neural networks for fast perception (System 1), HL systems for rules, safety, and local recovery (also System 1), and LLM agents providing high-level feedback and learning from the HL system's data (System 2). The core proposition is that in the era of capable coding agents, a significant portion of an AI's learned experience could be maintained as an auditable, evolving software system.

marsbit31 мин. назад

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

marsbit31 мин. назад

Your Claude Will Dream Tonight, Don't Disturb It

This article explores the recent phenomenon of AI companies increasingly using anthropomorphic language—like "thinking," "memory," "hallucination," and now "dreaming"—to describe machine learning processes. Focusing on Anthropic's newly announced "Dreaming" feature for its Claude Agent platform, the piece explains that this function is essentially an automated, offline batch processing of an agent's operational logs. It analyzes past task sessions to identify patterns, optimize future actions, and consolidate learnings into a persistent memory system, akin to a form of reinforcement learning and self-correction. The article draws parallels to similar features in other AI agent systems like Hermes Agent and OpenClaw, which also implement mechanisms for reviewing historical data, extracting reusable "skills," and strengthening long-term memory. It notes a key difference from human dreaming: these AI "dreams" still consume computational resources and user tokens. Further context is provided by discussing the technical challenges of managing AI "memory" or context, highlighting the computational expense of large context windows and innovations like Subquadratic's new model claiming drastically longer contexts. The core critique argues that this strategic use of human-centric vocabulary does more than market products; it subtly reshapes user perception. By framing algorithms with terms associated with consciousness, companies blur the line between tool and autonomous entity. This linguistic shift can influence user expectations, tolerance for errors, and even perceptions of responsibility when systems fail, potentially diverting scrutiny from the companies and engineers behind the technology. The article concludes by speculating that terms like "daydreaming" for predictive task simulation might be next, continuing this trend of embedding the idea of an "inner life" into computational processes.

marsbit33 мин. назад

Your Claude Will Dream Tonight, Don't Disturb It

marsbit33 мин. назад

Торговля

Спот
Фьючерсы
活动图片