TLDR: Anthropic has emerged as the leading large language model (LLM) provider in the enterprise market, capturing a 32% usage share and surpassing OpenAI, which now holds 25%. This shift is attributed to enterprises prioritizing reliability, safety, and performance, particularly in coding tasks, and is supported by a significant increase in enterprise AI budgets, which have more than doubled to $8.4 billion by mid-2025.
A new report from Menlo Ventures, released on July 31, 2025, reveals a significant shift in the enterprise large language model (LLM) market: Anthropic has overtaken OpenAI as the dominant provider. Anthropic now commands 32% of enterprise LLM usage, while OpenAI, which held a 50% share in 2023, has seen its share decline to 25%. Google has also made substantial gains, securing 20% of the market with its Gemini models, followed by Meta’s Llama at 9% and DeepSeek at 1%.
This market realignment coincides with a dramatic increase in enterprise spending on LLM technologies, which has more than doubled in just six months, rising from $3.5 billion in late 2024 to $8.4 billion by mid-2025. The report, based on a survey of 150 technical leaders across AI startups and large enterprises, indicates a maturing market where performance and operational outcomes are paramount.
Enterprises are increasingly favoring Anthropic’s Claude models for their reliability, safety, and consistent performance, especially in critical coding tasks. Anthropic’s models now account for an impressive 42% of usage in enterprise code-related workloads, more than double OpenAI’s 21% in the same segment. This surge in Anthropic’s adoption followed the release of Claude 3.5 Sonnet in June 2024, Claude 3.7 Sonnet in February 2025, and further solidified with Claude 4 and Claude Code in the second quarter of 2025.
Tim Tully, a partner at Menlo Ventures, commented on the shift: “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 report also highlights a broader trend: the focus of enterprise AI workloads has shifted predominantly to inference rather than training. By mid-2025, 74% of AI startups and 49% of large enterprises are using models almost exclusively for inference. This indicates a move away from internal model training, with only 9% of teams reporting investments in internally trained models. The market is increasingly relying on powerful, frequently updated general-purpose models and APIs from providers.
Also Read:
- Anthropic’s Valuation Soars Amidst Significant Funding Rounds and Revenue Growth
- Anthropic Revolutionizes Finance with New AI-Powered Analysis Solution
Furthermore, the report notes that while vendor switching is rare (only 11% of teams changed providers in the past year), a significant 66% upgraded to newer models from their existing vendors. This underscores the importance of continuous improvement and timely releases in maintaining market relevance. Anthropic’s success is attributed not only to model quality but also to its ability to build a robust ecosystem of tools, ensure API compatibility, focus on compliance, and offer competitive pricing, integrating well into existing enterprise toolchains and earning trust among various stakeholders.
