SambaNova raises $1B Series F at $11B valuation; JPMorgan deploys inference chips on-prem
SambaNova, an AI inference chip startup, closed the first tranche of its Series F round at $1 billion, reaching an $11 billion post-money valuation. General Atlantic led the round with participation from Seligman Ventures, T. Rowe Price Associates, Capital Group, Intel Capital, BlackRock, QFO Capital, Vista Equity Partners, Cambium Capital, Battery Ventures, Assam Ventures, and Qatar Investment Authority. Additional investors are expected in a second close.
Alongside the funding, JPMorgan Chase selected SambaNova as an inference-infrastructure partner, deploying its SN40L and SN50 systems for on-premises AI inference on enterprise AI workloads. SambaNova's custom Reconfigurable Dataflow Unit (RDU) architecture is designed for single-rack inference deployment with ~10 kW power consumption, a fraction of GPU-based alternatives. The SN50 unveiled in February 2026 ships later in 2026, with SoftBank as its first deployment partner.
SambaNova's valuation has roughly quintupled in five months—from ~$2.2 billion at its February 2026 Series E to $11 billion today. This reflects surging investor demand for NVIDIA alternatives focused on inference, as enterprises and banks move to secure, on-prem AI deployment. The company was valued near $1.6 billion in late 2025 during failed Intel acquisition talks.
For practitioners, the signal is clear: investors are betting on a commodity-shift from NVIDIA-dominant training to heterogeneous inference stacks. On-prem deployment for sovereign and regulated-industry use cases (banking, government) is now a tier-one category that commands venture capital at scale-up valuations.
Sources
- Primary source
- generalatlantic.com
“SambaNova today announced it has completed the first close of $1 billion in strategic financing as part of a Series F round, valuing the company at $11 billion post money”
- techcrunch.com
“SambaNova said it has been selected by JPMorganChase as an inference-infrastructure partner, with its SN40L and SN50 systems set to power secure, on-premises AI inference at the bank”