hn.today

Tenstorrent raises $1.5B [$6.2B valuation] to expand its open compute ecosystem

siliconangle.com6 points1 comments
Screenshot of Tenstorrent raises $1.5B [$6.2B valuation] to expand its open compute ecosystem

Tenstorrent raised $1.5 billion in a Series E round at a $6.2 billion valuation, led by Trace Capital and Innovation X with participation from Protagonist, Morgan Stanley, Fidelity, ARK Invest and T. Rowe Price. Led by chip designer Jim Keller, the company evolved from silicon design into a full AI compute platform vendor offering Galaxy servers, liquid-cooled TT-QuietBox workstations and Blackhole PCIe accelerator cards. Its hardware portfolio centers on Tensix Neo accelerators and Ascalon CPUs built on the open-standard RISC-V ISA, paired with GDDR6 memory and scalable interconnects so systems can be networked from single workstations to massive superclusters without excessive power draw.

Tenstorrent ships a complementary open-source software stack - TT-Forge (a compiler bridging PyTorch to its RISC-V chips), TT-Metalium (an SDK for hardware optimization) and TT-LLK (low-level kernel tools) - to simplify model compilation, tuning and execution on its silicon. Galaxy servers are in production use by customers like Virtu Financial and available through colocation partners such as Equinix. The new funds will expand Galaxy production, global deployments, RISC-V research and hiring. Investors and analysts highlight the integrated chip-plus-software strategy and reported video-generation benchmark performance as evidence Tenstorrent is positioning itself as a viable, more open and potentially lower-cost alternative to dominant GPU suppliers.

Read on siliconangle.com1 comments on Hacker News

Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.

More in AI

Evals for 3D CAD

Evals for 3D CAD

MUSE is a new benchmark for evaluating text-to-CAD models on complex assembly tasks, focusing on functionality, manufacturability, and assemblability. Experiments reveal current models struggle with producing engineering-ready designs, highlighting the need for improved AI tools in industrial product design. (emergentmind.com)

The daily digest

Today's best Hacker News stories, summarized and screenshotted, one email a day.