Windsurf

In 2025, Windsurf's revenue reached $82M. The company previously reported $100M in 2025. Since its launch in 2021, Windsurf has shown consistent revenue growth.

Access: All users Available for unlimited use to all users is a fast, high-quality Devin Desktop Chat model based on Meta’s[Llama 3.1 70B](https://ai.meta.com/blog/meta-llama-3-1/). This model is optimized for speed, and is the fastest model available in Devin Desktop Chat. This is all while still being extremely accurate.

Windsurf (formerly Codeium) raised approximately $243M in total funding since its 2021 founding, with its last disclosed round being a $150M Series C led by General Catalyst in August 2024 at a $1.25B valuation, up from a $500M valuation at its Series B in January 2024. Notable investors include General Catalyst, Kleiner Perkins, and Greenoaks.

Sacra estimates [Codeium](https://sacra.com/c/codeium/) hit $82M ARR in July 2025, up from $40M in February 2025, at the time of its acquisition by [Cognition](https://sacra.com/c/cognition/).

OpenAI had been in talks to acquire Windsurf for about [$3 billion](https://www.cnbc.com/2025/04/16/openai-in-talks-to-pay-about-3-billion-to-acquire-startup-windsurf.html) in April, but the deal fell apart, and Google said on Friday that it hired Windsurf's co-founder and CEO Varun Mohan. Google is paying $2.4 billion in licensing fees and for compensation, as CNBC [previously reported](https://www.cnbc.com/2025/07/11/google-windsurf-ceo-varun-mohan-latest-ai-talent-deal-.html).

Windsurf, the company formerly known as Codeium, has been in talks to raise fresh funds at a[ $2.85 billion valuation](https://techcrunch.com/2025/02/19/ai-coding-startup-codeium-in-talks-to-raise-at-an-almost-3b-valuation-sources-say/) led by Kleiner Perkins, TechCrunch reported in February. The company has reached about $40 million in annualized recurring revenue (ARR), according to our reporting. That revenue run rate is much lower than Cursor’s, which reportedly makes $200 million on an ARR basis. Cursor has been in talks to raise capital at about[ $10 billion valuation](https://techcrunch.com/2025/03/07/cursor-in-talks-to-raise-at-a-10b-valuation-as-ai-coding-sector-booms/), Bloomberg reported last month.

The startup's investors include Founders Fund, General Catalyst, Greenoaks and Kleiner Perkins. [TechCrunch](https://techcrunch.com/2025/02/19/ai-coding-startup-codeium-in-talks-to-raise-at-an-almost-3b-valuation-sources-say/) reported in February that Windsurf was raising a funding round at a $2.85 billion valuation.

Windsurf, the maker of a popular AI coding assistant, is in talks to be acquired by OpenAI for about $3 billion, Bloomberg [reported](https://www.bloomberg.com/news/articles/2025-04-16/openai-said-to-be-in-talks-to-buy-windsurf-for-about-3-billion).

Windsurf achieves 50% cost savings using Crusoe Cloud for their GPU infrastructure

We are both proud and excited to announce our Series C, $150 million raised at a $1.25 billion valuation, led by existing investor General Catalyst and with participation from existing investors Kleiner Perkins and Greenoaks.

We are both proud and excited to announce our Series C, $150 million raised at a $1.25 billion valuation, led by existing investor General Catalyst and with participation from existing investors Kleiner Perkins and Greenoaks.

- Chat: A whole different situation. The reality is that really large foundational models, whether closed-source or open-source, are just fine for Chat. In fact, given the breadth of questions that could appear, our best strategy to drive the most value is to provide maximum optionality to the user between models so that they could compare results from multiple models and use the best results. We provide smaller, Codeium-optimized Chat models for easier questions where faster results are better, but also provide optionality to Open AI’s GPT-4o, Meta’s LLama 3.1 405B, and very soon, Google’s Gemini and Anthropic’s Claude Sonnet. Note that we do not expose every OpenAI model or every Google model or every Anthropic model - just the best and biggest. Latency is not as strong of a consideration for Chat so we should always be suggesting the best models, not ones that are objectively suboptimal choices.

Today, that feels less of a case. Not only are there many reports of other closed-source models such as Anthropic’s Claude Sonnet outperforming OpenAI’s GPT-4o model on a variety of tasks, but open source models such as Meta’s Llama 3.1 405B and NVIDIA’s Nemotron 340B have reached a comparable level of quality and opened a whole new world of possibility. Because these models are open-sourced, anyone has the potential to do additional training on these models in order to clearly outperform close sourced models such as GPT-4o on application-specific tasks. And because these models are commercially usable, there is now reason to invest in doing this additional model tuning now, unlike before when the exhorbitant 8 or 9 figure price tag for pretraining such a model completely eliminated the possibility of owning a massive proprietary model. In many ways, the rise of OpenAI competitors and open-source models have democratized the top-end models.

- Chat: A whole different situation. The reality is that really large foundational models, whether closed-source or open-source, are just fine for Chat. In fact, given the breadth of questions that could appear, our best strategy to drive the most value is to provide maximum optionality to the user between models so that they could compare results from multiple models and use the best results. We provide smaller, Codeium-optimized Chat models for easier questions where faster results are better, but also provide optionality to Open AI’s GPT-4o, Meta’s LLama 3.1 405B, and very soon, Google’s Gemini and Anthropic’s Claude Sonnet. Note that we do not expose every OpenAI model or every Google model or every Anthropic model - just the best and biggest. Latency is not as strong of a consideration for Chat so we should always be suggesting the best models, not ones that are objectively suboptimal choices.

Today, that feels less of a case. Not only are there many reports of other closed-source models such as Anthropic’s Claude Sonnet outperforming OpenAI’s GPT-4o model on a variety of tasks, but open source models such as Meta’s Llama 3.1 405B and NVIDIA’s Nemotron 340B have reached a comparable level of quality and opened a whole new world of possibility. Because these models are open-sourced, anyone has the potential to do additional training on these models in order to clearly outperform close sourced models such as GPT-4o on application-specific tasks. And because these models are commercially usable, there is now reason to invest in doing this additional model tuning now, unlike before when the exhorbitant 8 or 9 figure price tag for pretraining such a model completely eliminated the possibility of owning a massive proprietary model. In many ways, the rise of OpenAI competitors and open-source models have democratized the top-end models.

Two years ago, when ChatGPT came out, it seemed that OpenAI had an insurmountable lead when it came to these foundational models. No matter what competitors or open source did, it always felt that the most recent GPT model was clearly the gold standard of achieved model potential.

Today, we are excited to announce our $65m Series B round at a $500m valuation, led by Kleiner Perkins and with participation from repeat investor Greenoaks as well as General Catalyst.