Low profile, half-length PCIe Gen 4 board with a single Grayskull™ processor, our limited-availability entry-level board operating at 75W

The Grayskull™ DevKit offers a range of capabilities, tailored to suit the needs of ML developers seeking alternatives to traditional GPUs. Our hardware supports a wide array of models, ensuring versatility in your AI projects.

TT-Buda: run models right away

Great for production customers who want to get models up and running with ease. They want flexibility, but don’t have time to program new operations or contact support every time something changes.

TT-Metalium: Open Access to Tenstorrent Hardware & Software

Great for development customers who want to customize their models, write new ones or even run non-machine learning code. No black boxes, encrypted APIs or hidden functions.

While Grayskull™ supports a wide range of models, here are five key models that highlight its capabilities:

• BERT: Ideal for natural language processing tasks.
• ResNet: A classic choice for image recognition.
• Whisper: Exceptional for speech recognition and translation.
• YOLOv5: Excellent for real-time object detection.
• U-Net: Highly effective for image segmentation.

For a comprehensive list of all supported models, please visit our full model support page.

• Architecture: Tenstorrent Grayskull™
• Tensix Cores: 96
• AI Clock: 1 GHz
• SRAM: 96MB
• Memory: 8GB LPDDR4 @ 102.4 GB/sec
• System Interface: PCIe 4.0 x16
• Total Board Power: 75W
• Dimensions (w/o Cooling Kit): 18mm x 167.5mm x 69mm
• Dimensions (w/ Cooling Kit): 18mm x 257mm x 98mm

• CPU:  x86_64 architecture
• Motherboard: PCI Express 4.0 x16 slot, single width
• Memory: 64GB
• Storage: 100GB (≥2TB recommended)
• Power Connectors:PCIe 6-pin (required for powering fan)
• Operating System: Ubuntu version 20.04 (Focal Fossa)
• Internet Connection: Required for driver and stack installation

Standard height, 3/4 length PCIe Gen 4 board with one Grayskull™ processor, our solution for balancing power and throughput, operating at up to 200W

The Grayskull DevKit offers a range of capabilities, tailored to suit the needs of ML developers seeking alternatives to traditional GPUs. Our hardware supports a wide array of models, ensuring versatility in your AI projects.

TT-Buda: run models right away

Great for production customers who want to get models up and running with ease. They want flexibility, but don’t have time to program new operations or contact support every time something changes.

TT-Metalium: Open Access to Tenstorrent Hardware & Software

Great for development customers who want to customize their models, write new ones or even run non-machine learning code. No black boxes, encrypted APIs or hidden functions.

While Grayskull supports a wide range of models, here are five key models that highlight its capabilities:

• BERT: Ideal for natural language processing tasks.
• ResNet: A classic choice for image recognition.
• Whisper: Exceptional for speech recognition and translation.
• YOLOv5: Excellent for real-time object detection.
• U-Net: Highly effective for image segmentation.

For a comprehensive list of all supported models, please visit our full model support page.

• Architecture: Tenstorrent Grayskull™
• Tensix Cores: 120
• AI Clock: 1.2 GHz
• SRAM: 120MB
• Memory: 8GB LPDDR4 @ 118.4 GB/sec
• System Interface: PCIe 4.0 x16
• Total Board Power: 200W
• Dimensions (w/o Cooling Kit): 36mm x 260mm x 111mm
• Dimensions (w/ Cooling Kit): 36mm x 399mm x 114mm

• CPU:  x86_64 architecture
• Motherboard: PCI Express 4.0 x16 slot, dual width
• Memory: 64GB
• Storage: 100GB (≥2TB recommended)
• Power Connectors: PCIe 6+2-pin and PCIe 6-pin
• Operating System: Ubuntu version 20.04 (Focal Fossa)
• Internet Connection: Required for driver and stack installation