SambaNova raises $1B at $11B valuation, quintuples worth in months as AI chip race intensifies

SambaNova Secures $1 Billion Series F Funding at $11 Billion Valuation

AI chip startup SambaNova Systems has completed the first close of a $1 billion funding round at an $11 billion valuation, marking one of the most dramatic revaluations in the semiconductor sector this year1

1

. The Series F funding round was led by General Atlantic, with significant participation from Seligman Ventures, T. Rowe Price Associates, and Capital Group5

5

. The company expects a second close within weeks as additional investors join, according to CEO and co-founder Rodrigo Liang1

1

.

The SambaNova valuation jump is particularly striking given that the Palo Alto-based company was worth roughly $2 billion earlier this year, meaning it has quintupled its worth in a matter of months4

4

. This latest round comes just five months after SambaNova closed a $350 million Series E in February alongside the unveiling of its SN50 chip1

1

. The speed of this re-rating reflects intense investor appetite for any credible Nvidia alternative as AI spending shifts from training models to running them4

4

.

JPMorgan Chase Partnership Signals Enterprise Shift to On-Premise AI

Alongside the funding announcement, SambaNova revealed that JPMorgan Chase has selected it as an inference infrastructure partner, deploying SN40L and SN50 systems for secure, on-premise AI inference at the bank1

1

. Liang described the JPMorgan win as "a big deal" that sends a message to the banking industry about moving away from complete dependence on cloud services1

1

.

"For banks and for other industries where data is incredibly important, bringing this infrastructure on prem, bringing this infrastructure with models that are then under your control with your private data, and having all within your firewalls is an incredibly important aspect of running AI in a very secure and private manner," Liang explained5

5

. Banks of JPMorgan's caliber are now building their own private, secure AI infrastructure to run inference on their most sensitive models, a trend Liang expects to resonate beyond banking into other enterprises and governments1

1

.

AI Chip Startup Demonstrates Performance Edge with Heterogeneous AI Inference

SambaNova's technology combines its proprietary Reconfigurable Dataflow Unit (RDU) accelerators with Nvidia GPUs to create a heterogeneous inference platform that significantly outperforms GPU-only systems2

2

. Recent benchmarks conducted by Artificial Analysis showed SambaNova's SN50-series accelerators churning out 763 tokens per second in MiniMax M2.7 at short context lengths of 10,000 input tokens—several times faster than competing inference providers running on Nvidia GPUs alone2

2

.

The system accomplishes this by disaggregating the inference pipeline: computationally intensive prefill operations are handled by four Nvidia H200 GPUs, while memory-bandwidth-bound decode operations run on a single SambaNova rack containing 16 SN50 accelerators2

2

. Liang told Bloomberg that SambaNova's chips can run the decode portion of AI inference five to ten times faster than competing approaches5

5

. The AI chip startup positions itself as offering "premium inference" capable of running multi-trillion-parameter models at scale, fitting them onto a single rack to maximize speed1

1

.

Intel Partnership Deepens as AI Hardware Market Heats Up

SambaNova's relationship with Intel, a backer since its Series C and a participant in this latest funding round, has grown substantially1

1

. The companies announced a multi-year partnership in February to support AI inference development based on Intel's Xeon chip, and they now co-develop products and bring them to market together1

1

. "That gives us a great relationship with them that lets us leverage the scale of Intel with the technology we have," Liang said1

1

.

The partnership is notable given that SambaNova's chairman is Lip-Bu Tan, who also serves as chief executive of Intel4

4

. Prior acquisition discussions between the two companies had broken down, with a December report from Bloomberg News indicating Intel had considered acquiring SambaNova at a valuation of roughly $1.6 billion1

1

. Regulators granted antitrust approval for the investment arrangement in May5

5

.

Supply Chain Focus and IPO Prospects for 2027

SambaNova plans to use the funding proceeds to scale the business and secure its supply chain against what Liang described as "an incredible wave of demand." "We're using that capital to secure the supply chain," he said, describing it as essential to fulfilling orders and buying the materials needed to deliver over the next 12 months1

1

. The company will also expand production capacity and continue investing in chips, systems, and software for enterprise, sovereign AI, and neo-cloud customers5

5

.

When asked whether the Series E and F funding rounds meant the nine-year-old company had settled on staying independent, Liang remained noncommittal, noting that SambaNova continues to field acquisition interest1

1

. However, he indicated that momentum and growth will most likely drive the company toward "being public at some point." Liang told CNBC the company is considering an IPO in 2027, most likely in the U.S.5

5

Broader Market Context and Competition

The SambaNova funding round reflects a broader market dynamic where buyers want more than one company setting the price for AI infrastructure as spending shifts from training models to running them4

4

. Public market investors have shown bullish sentiment toward the semiconductor sector, with the PHLX semiconductor index up around 80% this year3

3

. Private markets have seen a flurry of activity around chip companies challenging incumbents, including South Korean startup Rebellions gearing up for an IPO on the Kospi in the first or second quarter of 2027, and Nvidia's deal to license technology from inference chip startup Groq3

3

.

SambaNova faces competition from a cluster of specialist firms including Cerebras and Groq, each arguing that its particular design suits the economics of inference better than general-purpose Nvidia GPUs4

4

. The company's SN50 chip, unveiled in February 2026, is due to begin shipping to customers in the second half of 2026, with SoftBank as its first deployment partner1

1

. SambaNova also counts Saudi Aramco, Intel, and various Japanese firms among its customers, targeting three main customer segments: sovereign clouds where governments fund local partners to build private clouds, neoclouds, and enterprises building for their own use1

1

.