Lin Qiao, CEO & Co Founder, Fireworks AI, on Centre stage during day three of Web Summit 2024 at the MEO Arena in Lisbon, Portugal. ” width=”970″ height=”647″ data-caption=’According to Lin Qiao, the future of A.I. lies in millions of specialized models built on proprietary data. <span class=”lazyload media-credit”>Sam Barnes/Sportsfile for Web Summit via Getty Images</span>’>
The soaring demand for A.I. has given rise to a new category of digital utility companies that sell compute power, access to models and developer infrastructure. Among the leaders of this pack is Fireworks AI, co-founded by former Meta executive Lin Qiao, who led the creation of PyTorch, a popular open-source machine learning framework, and a team of engineers from Meta and Google.
Fireworks AI is a platform for developers to build products faster and at lower cost than with proprietary models, using open-source models. It has access to many of the most capable open-source models on the market, such as Meta’s Llama series, Mistral, Qwen and DeepSeek. It also allows enterprises to upload their own data to train and fine-tune these models. Its clients include Cursor, Harvey, Uber and Shopify, among others.
Lin describes Fireworks as “a specialized intelligence platform,” as opposed to general intelligence. Specialized intelligence was what A.I. researchers primarily relied on before general intelligence became viable. “Before generative A.I. was a thing, there was no foundation model holding world knowledge together. GenAI changed that,” Lin explained to Observer. “Now, foundation models learn from the public internet and massive labeled datasets, creating a deeper, more generalized knowledge base that you could use directly as a black-box API.”
But Lin believes that, amid the abundance of public data and general intelligence models, the most valuable uses of A.I. will, counterintuitively, come from specialization.
“Because foundational models do not have access to the private data locked inside applications and enterprises,” she said. “The majority of data is private, locked inside enterprises as proprietary IP and information that would never get shared outside the company.”
Training and fine-tuning models with that private data creates an ongoing need for Fireworks’ services. “This is a continuous process because applications keep evolving, data distribution changes, and base models keep improving,” Lin said. “We have customers tuning once a week, once a day, or even once every few hours.” She predicted that this tuning process would soon be fully automated.
Once a model is finely tuned, Fireworks helps optimize it for inference speed and cost. The company offers some of the fastest inference—the speed at which an A.I. generates a response—in the industry. For example, Cursor’s code editor uses Fireworks’s speculative decoding to deliver code suggestions up to 13 times faster than traditional setups.
Fireworks processes more than 30 trillion tokens in daily inference traffic (excluding training), more than OpenAI and Google’s Gemini, according to the latest published data.
The company makes money by charging users a flat rate per million tokens. Tokens are the basic unit of data that an A.I. reads, processes and generates; in English, a token is roughly four characters, or about three-quarters of a word.
“We provide one platform covering the whole end-to-end spectrum of model development, from quality to speed and cost. The end result is our customer gets better quality, much faster speed, and five to ten times lower cost, allowing them to go to production at a massive scale quickly,” Lin said.
The new moat
These days, A.I. executives like to talk about “moat,” or a competitive edge that allows a company to stay ahead of the competition. In a time when it’s easier than ever to turn an idea into an application thanks to A.I. coding tools, the traditional moat of products disappears.
“Data is the moat, because it cannot be copied,” Lin declared. “The data collected to understand user intent, user preferences and user engagement—what works well, what doesn’t work well and where you should optimize—is all your proprietary information, and that creates the asymmetry needed to compete. Whoever can turn this data into their proprietary intelligence can build on top of that. And that can compound.”
Fireworks competes with both closed-model providers (such as OpenAI, Anthropic and Google) and infrastructure platforms like Together AI, Replicate and AWS Bedrock. Its differentiation lies in focusing on open models while tightly integrating training, fine-tuning and high-performance inference into a single system.
“We don’t need a Ferrari for grocery shopping.”
Besides the data moat, another argument for open models is unit economics. By allowing developers to choose from a wide range of open-weight models, platforms like Fireworks can match each task with the most cost-efficient level of intelligence. This flexibility is increasingly important as companies look to deploy A.I. at scale. Using a single, frontier model for every task quickly becomes prohibitively expensive.
“We don’t need to drive a Ferrari to go grocery shopping,” Lin said. “There are so many tasks we solve day-to-day at varying levels of complexity. Some are extremely hard, requiring beyond-human-level intelligence to solve. Others are not that hard. If you use a vendor who can help you automatically select the best model suitable for solving a particular task, you get the quality you need at the lowest cost.”
When Lin founded Fireworks two years ago, the company initially focused on inference, treating it as “one size fits one.” Now, it is doubling down on training as well, driven by the rapid improvement and release cadence of open models. Open model quality has significantly narrowed the gap with closed models, while release cycles have accelerated from monthly to weekly. New models frequently top benchmarks and approach frontier-level performance.
“This makes training particularly appealing. With your private data and a little bit of tuning, you can stay on top,” Lin said.
She continued to conclude, “We believe specialized and generalized intelligence will coexist, but the world will not be dominated by a few generalized models. There will be millions of specialized intelligence models—one per use case.”
