Skip to main content

Etched Sohu ASIC

Vendor: Etched

Architecture: Transformer ASIC

Category: ASIC Dedicated Accelerator

Introduction

An ASIC chip hard-coded specifically for the Transformer architecture. It removes redundant logic used for graphics rendering and general-purpose computing in GPUs, claiming to be several orders of magnitude faster than the NVIDIA H100 on large models such as Llama.

Specifications

ModelComputeMemoryInterfaceTDPProcess
SohuTransformer-dedicated (multiple times H100)External HBMPCIe 5.0TBATBA

Official Website

Visit Official Website

OS Support

WindowsLinuxmacOSAndroid
✅ (Etched Cloud API)

Version History

VersionRelease DateDescription
Sohu Pre-release2025Claims Llama inference speed far exceeds H100

Performance Benchmarks

ModelTaskPerformance Metric
SohuLlama 2 70B InferenceUltra-high throughput (official data)
SohuTransformer InferenceDedicated Transformer inference acceleration

Pricing

ModelReference PriceNotes
SohuCloud APIEtched Cloud API
SohuContact vendorEnterprise deployment

Quick Installation

Etched Cloud (API)

pip install etched-sdk

Sohu is a dedicated Transformer ASIC, supporting only Transformer architecture model inference.

Code Examples

Python (Etched API)

from etched import EtchedClient

client = EtchedClient(api_key="your-key")
response = client.generate(
model="llama-3-70b",
prompt="Hello",
max_tokens=100
)

Architecture Highlights

  • Transformer ASIC: The world's first ASIC designed specifically for the Transformer architecture, implementing attention mechanisms directly in hardware
  • Extreme Inference: Bypasses the flexibility overhead of general-purpose computing, pushing Transformer inference efficiency to the limit
  • Limitations: Only supports Transformer architecture models; does not support CNNs, graph neural networks, etc.

Model Compatibility

Model/FrameworkSupportNotes
Transformer LLM✅ NativeLlama/GPT/Qwen etc.
CNN ModelsNot supported
Non-TransformerTransformer architecture only

If you're evaluating alternatives, the following products may also fit your scenario: