SambaNova

SambaNova Systems was founded in 2017 by Rodrigo Liang, Kunle Olukotun, and Christopher Ré, with Olukotun and Ré both Stanford computer science professors with deep expertise in computer architecture and machine learning respectively, building a specialized AI chip architecture and accompanying software stack designed as an alternative to Nvidia's dominant GPU-based approach to AI computing infrastructure, arguing that a chip architecture designed specifically around the computational patterns of modern AI workloads, rather than adapted from a chip architecture originally designed for computer graphics rendering, could offer meaningful performance and efficiency advantages for AI training and inference workloads. SambaNova has positioned its offering as a full-stack AI infrastructure platform rather than only a chip product, providing enterprises and government customers with an integrated hardware and software system intended to simplify deploying and running large AI models, including both training custom models and running inference on pre-trained models, an approach the company argues addresses the significant complexity many enterprises face trying to assemble a complete, properly optimized AI computing stack using separately sourced hardware and software components from many different vendors. The company has particularly emphasized sovereign and on-premises AI deployment as a key market segment, targeting government agencies, financial institutions, and other enterprises with data sovereignty or security requirements that necessitate running AI workloads on infrastructure they directly control rather than relying entirely on cloud-based AI services from major providers, a positioning that has resonated particularly with government customers in various countries seeking to build independent AI computing capacity not dependent on foreign cloud providers. SambaNova has navigated the significant competitive challenge shared by all AI chip startups seeking to challenge Nvidia's dominant market position, given Nvidia's extensive software ecosystem and continued architectural improvements that have made it difficult for alternative chip architectures to gain substantial market share despite various startups' claimed performance advantages for specific workloads. The company has raised very substantial venture funding from investors including SoftBank Vision Fund, BlackRock, and Intel Capital, reaching a valuation exceeding $5 billion.

News

10/13/2026

SambaNova Says AI Shift Focuses on Inference, Efficiency and ROI

SambaNova's CEO says the AI infrastructure boom is moving toward inference-focused, efficiency and ROI-driven deployments, prompting enterprises to run private systems behind firewalls and signaling coexistence between frontier models and private AI.

Bloomberg

7/11/2026

SambaNova Systems Secures $1 Billion in Series F Funding

This funding round elevates SambaNova's valuation to $11 billion, reinforcing its position in the AI infrastructure market and supporting the development of advanced AI chips.

Crunchbase News

7/9/2026

DigiTech ASEAN Thailand & AI Connect 2026 to Showcase AI Innovations

SambaNova Systems, as a Silver Sponsor, will present its solutions at a key event connecting over 12,000 tech professionals in Bangkok, enhancing its visibility in the ASEAN digital landscape.

PR Newswire

7/8/2026

SambaNova Systems raises $1 billion, achieving $11 billion valuation

SambaNova Systems secures $1 billion in Series F funding to enhance AI infrastructure, reflecting strong investor confidence and growth potential.

Yahoo Finance

7/8/2026

SambaNova raises $1 billion at $11 billion valuation

The late-stage $1 billion round, led by General Atlantic, values the AI chip maker at $11 billion and signals investor confidence in demand for high-speed AI inferencing and infrastructure to challenge Nvidia.

Bloomberg

Bloomberg

TechCrunch