TLDR: A new report from Menlo Ventures reveals a significant shift in the enterprise AI market, with Anthropic now holding 32% of the large language model (LLM) market share by usage, surpassing OpenAI, which has fallen to 25%. This change is largely attributed to Anthropic’s Claude Sonnet models, known for their performance in production environments and coding tasks. The enterprise AI market has seen rapid growth, doubling to $8.4 billion in the last six months.
A recent market report from Menlo Ventures, titled ‘2025 Mid-Year LLM Market Update,’ indicates a substantial shakeup in the enterprise AI landscape. Anthropic has emerged as the new leader in the enterprise large language model (LLM) market, capturing 32% of the market share by usage. This marks a significant shift, as OpenAI, which previously held a dominant 50% share through 2023, has now seen its share decline to 25%. Google is also a notable player, increasing its market share to 20% with improvements in its Gemini platform, while Meta’s Llama holds 9% and DeepSeek accounts for 1%.
Anthropic’s ascent is primarily attributed to its enterprise-focused Claude Sonnet models, specifically the Claude 3.5 Sonnet released in June 2024 and Claude 3.7 Sonnet introduced in February 2025. These models are praised for their extended context windows, reduced hallucinations, and superior performance, particularly in coding-related AI applications. Anthropic now commands an impressive 42% of the enterprise coding market share, more than double OpenAI’s 21%.
Tim Tully, a partner at Menlo Ventures, commented on this market evolution, stating, ‘Some might be surprised to see Anthropic overtake OpenAI, given its first-mover advantage. But our research puts real numbers behind what we’ve heard anecdotally from the market: Teams are prioritizing real performance in production.’
The enterprise AI market itself is experiencing explosive growth. In just the first half of 2025, enterprise spending on LLMs more than doubled, climbing from $3.5 billion at the start of the year to $8.4 billion currently. Menlo Ventures projects this figure to reach approximately $13 billion by the end of the year.
The report also notes a clear preference for closed models over open-source alternatives within the enterprise sector. More than half of enterprises do not utilize open-source models at all, with only 13% of daily enterprise workloads currently employing them, a decrease from 19% at the beginning of the year. Despite this, Meta’s Llama remains the leading open-source model, though its market share has also declined.
Also Read:
- AI Supremacy Race Intensifies: Tech Giants Pour Billions Amidst Bubble Concerns
- TELUS Leverages Anthropic’s Claude on Google Cloud to Propel Workplace Innovation and Engineering Excellence
While celebrating its market gains, Anthropic continues to navigate challenges, including an ongoing lawsuit concerning the use of copyrighted books to train its systems. The company is also reportedly in discussions to raise an additional $5 billion, which could push its valuation to $170 billion, underscoring investor confidence in its strategic direction and market position.
