ChatGPT SearchSep 28, 2026
Tenstorrent is pursuing a somewhat different strategy: general-purpose AI acceleration, an Ethernet-based scale-out architecture, and a comparatively open software stack. Its Blackhole Galaxy systems are now shipping
Data as of Oct 14, 2026Based on 42,870 AI responses
Reviewed by Dimitry Apollonsky
AI summary
Tenstorrent designs customizable RISC-V processors, neural processing unit IP cores, and hardware accelerators like Blackhole alongside its TT-Forge compiler. It comes up when buyers evaluate custom silicon designers, seek FLOPS-per-watt optimization for training runs, or source IP cores for neural processing units.
<1%
of 42,870 answers across Tenstorrent’s markets · since Jul 5
The market map
AI Semiconductor and Accelerator VendorsMentioned in
Share of answers in AI Semiconductor and Accelerator Vendors each week · 285 answers since Jul 5
Each engine’s share of its own answers across Tenstorrent’s markets · since Jul 5
Where Tenstorrent ranks in AI
Question: We are evaluating neural processing units (NPUs). Who designs custom AI silicon?
Google AI ModeSep 15, 2026
Tenstorrent, and SambaNova pioneer alternative dataflow and tensor-streaming architectures.
Question: Which startups are strongest for AI infrastructure that reduces GPU dependence?
ChatGPT SearchSep 20, 2026
Tenstorrent — A broader accelerator play rather than a pure inference ASIC.
tenstorrent.com 22%Other sites 78%
Excerpts where Tenstorrent appeared in the AI's answer
Tenstorrent, and SambaNova pioneer alternative dataflow and tensor-streaming architectures.
Tenstorrent: Building RISC-V-based AI and high-performance computing processors.
Excerpts where Tenstorrent appeared in the AI's answer
Tenstorrent is pursuing a somewhat different strategy: general-purpose AI acceleration, an Ethernet-based scale-out architecture, and a comparatively open software stack. Its Blackhole Galaxy systems are now shipping
Tenstorrent — A broader accelerator play rather than a pure inference ASIC.
Excerpts where Tenstorrent appeared in the AI's answer
Tenstorrent : Led by Jim Keller, their RISC-V processor architecture and software stack are explicitly built for scalable, flexible, and on-premise AI inference.
Excerpts where Tenstorrent appeared in the AI's answer
Tenstorrent – architecture focused on scalable matrix workloads and efficiency per watt
Tenstorrent — focuses on scalable, energy-efficient AI compute architectures
Excerpts where Tenstorrent appeared in the AI's answer
Tenstorrent Wormhole / Blackhole : Utilizes a flexible, RISC-V-based many-core architecture that excels at scaling out matrix math and managing distributed memory.