Blab AI
Baseten Secures $1.5 B Series F, Valuation Climbs to $13 B Amid AI Inference Boom
Baseten, the AI inference startup founded in 2019, is on the brink of closing a $1.5 billion Series F round that values the company at up to $13 billion, the Wall Street Journal reports. The deal is co‑led by Spark Capital, Sands Capital, Altimeter Capital and Wellington Management and features a split‑priced structure: some investors are buying at a $13 billion valuation, while others are paying at $11 billion.
This new infusion follows a $300 million Series E that Baseten closed just five months earlier at a $5 billion valuation. The jump to $13 billion represents a 160 % increase in worth in a remarkably short span, underscoring the surge of investor interest in AI inference infrastructure.
Baseten’s core product is a computational layer that processes model outputs after a user submits a prompt. The platform routes requests to the most cost‑effective model available, drawing from open‑source alternatives to OpenAI and Anthropic. By selecting the right model for each inference request, Baseten claims to cut inference costs while preserving performance—a feature that resonates as enterprises tighten their AI budgets.
The company’s rapid valuation climb mirrors a broader industry shift. Early funding focused on training large models, but capital is now moving toward inference—the phase where models generate predictions, text, images and agentic behaviors for end users. Analysts project that inference workloads will account for roughly two‑thirds of all AI compute by 2026, up from a smaller share in preceding years. Market research estimates the global AI inference infrastructure market at $28.7 billion in 2025, with expectations of reaching $89.4 billion by 2034.
Baseten’s investor roster includes a mix of venture capital firms and institutional money. Spark Capital, a generalist VC with a focus on enterprise software and infrastructure, is one of the lead participants. Sands Capital, Altimeter Capital and Wellington Management also co‑lead the round. Additional backers such as Battery Ventures, Blackbird, D.E. Shaw Ventures and Durable Capital Partners are reported to have joined, according to the Wall Street Journal.
The split‑priced structure indicates that while some investors are willing to pay the higher $13 billion valuation, others are comfortable at $11 billion, reflecting differing assessments of Baseten’s growth prospects. The deal’s size and valuation suggest that investors view Baseten as a key player in the emerging inference market, where companies are vying to provide the most efficient and cost‑effective inference services.
Baseten’s focus on routing to open‑source models also positions it against incumbents that rely heavily on proprietary models. By offering a layer that can switch between multiple models, Baseten enables enterprises to balance cost, performance and compliance requirements. The company’s strategy aligns with the broader industry trend of diversifying model portfolios to reduce dependence on a single vendor.
The company is headquartered in California, but its founders are Australians, a detail highlighted in a recent Reuters report that noted the startup’s international roots.
Looking ahead, Baseten will likely deploy the new capital to expand its inference platform, deepen integration with cloud providers and accelerate product development aimed at enterprise customers. The company has not yet announced specific product releases or timelines, and it remains unclear how quickly it will deploy the new funding.
In summary, Baseten’s $1.5 billion Series F and $13 billion valuation mark a significant milestone for an AI inference startup that has captured investor attention amid a surge in demand for cost‑effective inference solutions. The round positions Baseten to scale its platform and compete in a market where inference is becoming the most contested layer of AI infrastructure.
This new infusion follows a $300 million Series E that Baseten closed just five months earlier at a $5 billion valuation. The jump to $13 billion represents a 160 % increase in worth in a remarkably short span, underscoring the surge of investor interest in AI inference infrastructure.
Baseten’s core product is a computational layer that processes model outputs after a user submits a prompt. The platform routes requests to the most cost‑effective model available, drawing from open‑source alternatives to OpenAI and Anthropic. By selecting the right model for each inference request, Baseten claims to cut inference costs while preserving performance—a feature that resonates as enterprises tighten their AI budgets.
The company’s rapid valuation climb mirrors a broader industry shift. Early funding focused on training large models, but capital is now moving toward inference—the phase where models generate predictions, text, images and agentic behaviors for end users. Analysts project that inference workloads will account for roughly two‑thirds of all AI compute by 2026, up from a smaller share in preceding years. Market research estimates the global AI inference infrastructure market at $28.7 billion in 2025, with expectations of reaching $89.4 billion by 2034.
Baseten’s investor roster includes a mix of venture capital firms and institutional money. Spark Capital, a generalist VC with a focus on enterprise software and infrastructure, is one of the lead participants. Sands Capital, Altimeter Capital and Wellington Management also co‑lead the round. Additional backers such as Battery Ventures, Blackbird, D.E. Shaw Ventures and Durable Capital Partners are reported to have joined, according to the Wall Street Journal.
The split‑priced structure indicates that while some investors are willing to pay the higher $13 billion valuation, others are comfortable at $11 billion, reflecting differing assessments of Baseten’s growth prospects. The deal’s size and valuation suggest that investors view Baseten as a key player in the emerging inference market, where companies are vying to provide the most efficient and cost‑effective inference services.
Baseten’s focus on routing to open‑source models also positions it against incumbents that rely heavily on proprietary models. By offering a layer that can switch between multiple models, Baseten enables enterprises to balance cost, performance and compliance requirements. The company’s strategy aligns with the broader industry trend of diversifying model portfolios to reduce dependence on a single vendor.
The company is headquartered in California, but its founders are Australians, a detail highlighted in a recent Reuters report that noted the startup’s international roots.
Looking ahead, Baseten will likely deploy the new capital to expand its inference platform, deepen integration with cloud providers and accelerate product development aimed at enterprise customers. The company has not yet announced specific product releases or timelines, and it remains unclear how quickly it will deploy the new funding.
In summary, Baseten’s $1.5 billion Series F and $13 billion valuation mark a significant milestone for an AI inference startup that has captured investor attention amid a surge in demand for cost‑effective inference solutions. The round positions Baseten to scale its platform and compete in a market where inference is becoming the most contested layer of AI infrastructure.
