Tenstorrent Funding Reaches $1.5 Billion in Series E at $6.2 Billion Valuation

Tenstorrent announced on October 19, 2026, a $1.5 billion Series E round that values the company at $6.2 billion post-money. The round more than doubles the $2.6 billion valuation from its late-2024 Series D. Proceeds will fund volume production of Galaxy servers, next-generation hardware, and global expansion of the Santa Clara-headquartered firm.

Tenstorrent Funding Round Details and Investor Participation

The round was led by Trace Capital and Innovation X. New participants include ARK Invest, Fidelity, T. Rowe Price, Morgan Stanley, SanDisk, Cadence Ventures, Eclipse, and Tether Investments. Several Canadian investors also joined. The round nearly doubled the company’s prior valuation after it raised more than $693 million in the Series D.

Tenstorrent will use the capital to scale manufacturing of Galaxy servers built around Blackhole processors. Each 32-chip Galaxy system delivers 23 petaflops in FP8 precision and began commercial shipments in April 2026. Funds will also support development of the TT-Ascalon RISC-V CPU and Tensix AI cores, along with workforce growth across offices in Santa Clara, Austin, Toronto, Belgrade, Tokyo, and Bangalore.

The company offers hardware from individual cards to full racks and superclusters. It also licenses its RISC-V CPU and Tensix AI cores to partners such as Samsung and LG. This dual model of direct sales and IP licensing gives customers flexibility to deploy at different scales while generating recurring revenue for Tenstorrent.

Background on Open-Weight AI Chips and Market Position

Tenstorrent was founded in 2016 with origins in Toronto and now employs about 1,000 people, including roughly 200 in Canada. CEO Jim Keller, previously responsible for AMD Zen architecture, Apple A4/A5 chips, and Tesla’s FSD silicon, has steered the company toward an open RISC-V architecture paired with an open-source software stack that includes TT-Forge and TT-Metalium.

The approach targets efficient inference workloads for open-weight models, including large language models, coding assistants, and video generation tools. Tenstorrent positions its products as lower-cost, customizable alternatives to offerings from Nvidia and AMD. A May 2026 demonstration of Galaxy servers running DeepSeek models highlighted performance and cost advantages that preceded the current funding round.

Earlier in 2026, reports surfaced of takeover interest from Intel and Qualcomm at valuations up to $5 billion. Tenstorrent declined those approaches and instead pursued independent growth. The new capital and expanded investor base, which includes ARK Invest and Fidelity, reinforce that strategy while providing resources to compete in high-density inference deployments.

Future Plans and Practical Implications for AI Hardware Buyers

Tenstorrent plans continued scaling of Galaxy production lines, introduction of subsequent hardware generations, and broader commercial availability in Asia and Europe. The company will also expand IP licensing agreements that allow partners to integrate Tensix cores into their own designs.

For enterprise buyers and model developers, the funding signals greater supply of systems optimized for open-weight workloads. Organizations evaluating alternatives to dominant GPU vendors can now access production-grade hardware with documented 23 petaflops per rack and an open software stack that reduces vendor lock-in. Procurement teams should monitor Tenstorrent’s roadmap updates and licensing terms when planning 2027 inference capacity.

Investors gain exposure to a company with a clear path to volume shipments and diversified revenue from both hardware sales and IP. The participation of established firms such as Fidelity and T. Rowe Price alongside newer entrants like Trace Capital indicates sustained confidence in the open RISC-V segment of the AI hardware market.