Sail Research closes $80 million Series B funding round for max-efficiency AI agent infrastructure and inference optimization.

Sail Research, the infrastructure company purpose-built for long-horizon AI agents, has announced $80 million in Seed and Series A funding at a $450 million valuation. Kleiner Perkins led the Series A while Sequoia led the Seed round. Additional investors included Redpoint Ventures, Theory Ventures, Vine Ventures, CRV, A*, and Abstract Ventures, alongside angel investors John Hennessy (chairman of Alphabet Inc.), Lip-Bu Tan (CEO of Intel), and Tri Dao (Chief Scientist at Together AI).

The next generation of AI infrastructure must be built for agents operating autonomously over hours and days, not the brief turn-by-turn interactions that shaped today's platforms. While the underlying stack was optimized for responding to a human waiting at a prompt, it wasn't designed for the fundamental differences in how AI agents function. Agents are constrained only by available compute and context—the more they receive, the better they perform. With global AI spend projected to reach $2.5 trillion in 2026, the most ambitious agent workloads remain inaccessible to most organizations, limited not just by cost but by the rate limits and scale ceilings of platforms never designed for long-horizon use.

Sail Research was built to eliminate those constraints. The company provides the first infrastructure platform purpose-built for long-horizon AI agents, combining two core components: an inference stack rebuilt from scratch around throughput and efficiency, designed for agents consuming billions of tokens on a single task; and Sailboxes, a sandbox environment that runs for hours and days while charging only for time agents are actively working. Together, they provide the economics and scale for teams to build maximally ambitious agents without hitting the walls that stop most production deployments.

"Sail exists to make intelligence abundant," said Neil Movva, co-founder and CEO of Sail. "Every decision we make, from the chip level to the API, is about giving teams the tokens, the scale, and the runtime to build agents without limits."

Sail's efficiency advantage comes from proprietary infrastructure optimizations, including deep customization of open-source inference engines to push GPU performance toward frontier capabilities, intelligent workload distribution across providers for maximum resilience, and strategic use of underutilized compute. In a recent benchmark, Sail's inference topped BrowseComp-Plus, a leading deep research evaluation, achieving 90.72% accuracy at up to 10 times lower cost than leading alternatives.

"Most inference infrastructure was designed to minimize latency on a single request, but that's the wrong optimization for agents, which need to sustain throughput across thousands of concurrent calls over hours," said Samir Menon, co-founder and CTO of Sail. "We've rebuilt the stack around that constraint, and the efficiency gains compound across every layer."

Movva spent years at NVIDIA pushing GPU performance to its limits before building infrastructure expertise at Apple and Together AI. Menon, also from Apple, has experience building systems at massive scale.

"The infrastructure layer for the agent era is one of the most important bets in AI right now, and Neil and Samir are exactly the founders to build it," said Aditya Naganath, partner at Kleiner Perkins. "They bring a rare combination of deep compute expertise and systems rigor that only comes from having built at the limits of scale. Together, they're building the defining inference platform for long-horizon agents."

Sail's infrastructure already powers AI-driven workflows at Parallel Web Systems, Jack and Jill, and Detail.dev. "We and Sail share a belief that background agents are about to do far more useful work," said Travers Nisbet, co-founder of Parallel. "Getting there takes efficient, scalable inference paired with the highest-quality context, including from the web. Sail is building the inference side of that, and we're glad to be aligned on where this is going."

Detail.dev, a California-based code review platform, uses Sail to power agents that analyze pull requests and codebases at a scale and depth that previously required significant human engineering time. "Building on Sail lets us ship long-horizon agents with great economics," said Dan Robinson, CEO of Detail.dev. "Trillions of tokens and counting, we're happy customers."

Sail's API is compatible with existing OpenAI-based workflows and supports leading open-source models including DeepSeek, Gemma, GLM, Kimi, and Nemotron.