SambaNova Raises $1B at $11B Valuation for AI Inference Chips

SambaNova Systems secures major funding round

SambaNova Systems has closed a $1 billion financing round that values the AI chip startup at $11 billion, according to details first reported by CNBC. General Atlantic led the investment, with participation from Seligman Ventures, T. Rowe Price, and Capital Group.

The substantial raise follows an earlier funding round this year that brought in more than $350 million from investors including Intel, which also established a partnership with the company.

Why it matters

As enterprises move beyond AI experimentation to production deployments, the inference chip market has become a critical battleground. SambaNova's focus on on-premise installations addresses corporate concerns about data security and latency that cloud-based solutions can't fully resolve. The $11 billion valuation signals investor confidence that alternatives to Nvidia's dominant GPU architecture can capture meaningful market share, particularly in regulated industries like financial services where control over AI infrastructure is paramount.

Targeting the inference chip market

SambaNova is positioning itself within the growing field of AI inference semiconductors—chips designed to run large language models efficiently after they've been trained. The company's latest offering, the SN50 chip, ships as part of a complete server unit rather than as a standalone graphics processing unit like those Nvidia has built its business around.

The architectural difference reflects a strategic bet on workload-specific optimization. While Nvidia's GPUs have proven essential for training massive AI models, SambaNova argues that inference workloads—which involve running deployed models to generate responses—benefit from purpose-built hardware.

Enterprise deployments and JPMorgan partnership

A key differentiator for SambaNova is its emphasis on on-premise installations. Rather than routing inference requests through third-party cloud providers or AI labs, companies can deploy SambaNova's server units directly in their own data centers.

JPMorgan Chase announced Wednesday it would implement SambaNova's systems for on-premise inference across its enterprise AI workloads. The bank's adoption underscores demand from heavily regulated industries that require direct control over AI infrastructure for security and compliance reasons.

SambaNova has positioned on-premises inference as offering faster response times and enhanced security compared to cloud-based alternatives, since the company using the technology maintains full operational control.

Broader competitive landscape

The financing comes amid heightened activity in the semiconductor sector. The PHLX semiconductor index has climbed approximately 80% this year, reflecting investor enthusiasm for companies providing the fundamental infrastructure for AI deployment.

Private markets have seen parallel momentum. South Korean startup Rebellions announced plans for an initial public offering on the Kospi exchange in the first or second quarter of 2027, according to statements made by CEO Sunghyun Park. Last year, Nvidia itself signed a licensing agreement with inference chip startup Groq, acknowledging the specialized nature of inference workloads.

The competitive dynamics suggest the AI chip market is fragmenting beyond the training phase that made Nvidia dominant, with inference representing a distinct opportunity for specialized architectures.

These details were first reported by CNBC.