Jim Keller: ‘AI Still Obeys the Old Laws of Compute’
By Sally Ward-Foxton, EE Times | 06.25.2026
SANTA CLARA, Calif. — When EE Times visited Tenstorrent CEO Jim Keller’s office a year ago, the whiteboard outside his office door read: “We’re going to WIN!” On returning a year later, it reads: “Holy Shit, That’s Fast!”
In the aftermath of Tenstorrent’s TT-Deploy event, where the company showed initial demonstrations of what its chips can do when deployed at scale, Keller told EE Times that Tenstorrent can beat the performance of both GPUs and more specialized AI hardware with its BlackHole Galaxy server.
Keller believes AI inference is ultimately a networking and memory problem, and that Tenstorrent’s architecture is now proving that at scale.
At TT-Deploy, the company demonstrated performance across a range of workloads. For example, 16 Tenstorrent Galaxy servers (512 chips) can inference DeepSeek-671B at up to 350 tokens per second per user at batch 32.
To read the full article, click here
Related Semiconductor IP
- Peripheral Sensor Interface (PSI5) Host Controller
- Link Acceleration Unit
- 64-bit, RISC-V, ultra-high performance processors
- 64-bit, RISC-V, performance and data computation processors
- 32-bit, RISC-V, deeply embedded processors
Related News
- Mythic Acquires Videantis, One of Europe’s Leading Digital Processor IP Companies, to Build the World’s Most Energy-Efficient AI Compute Platform
- RaiderChip showcases the evolution of its local Generative AI processor at ISE 2026
- MIPS and INOVA Collaborate to put Physical AI into the palm of Robotic hands with new Reference Platform
- Cadence and NVIDIA Expand Partnership to Reinvent Engineering for the Age of AI and Accelerated Computing
Latest News
- EdgeAI Licensed Andes Technology CPU IP to Power Next-Generation Edge AI Neuromorphic Solution
- Jim Keller: ‘AI Still Obeys the Old Laws of Compute’
- OpenAI and Broadcom unveil LLM-optimized inference chip
- RAAAM Selects Avnet ASIC as its VCA Partner for TSMC’s 2nm GCRAM Development and Qualification
- IBM Debuts World’s First Sub-1 Nanometer Chip Technology