Skip to content
HyperAccel SDK Documentation
OpenAI 호환 API
한국어
English
Initializing search
GitHub
HyperAccel SDK Documentation
GitHub
Overview
Getting Started
Getting Started
Quick Install
Driver Installation and Setup
Driver Installation and Setup
Overview
Supported Environments
Installing
Verifying the Installation
Upgrading and Removing
Quickstart
Tutorials
Software
Software
Legato Language
Legato Language
Overview
Programming Guide
Programming Guide
Overview
Programming Model
PyTorch Integration
Compile PyTorch Model with Legato
Decorators And Sessions
Argument Binding
Compiler Session
Context Management
Layouts And Memory Kinds
Memory Operations
Synchronization
Types
Static Evaluation
Runtime Control Flow
Broadcasting
Bertha
Bertha
Bertha API
Broadcasting on Bertha
Bertha Layouts
Instruction Table ABI
API Reference
API Reference
Overview
Tensor Operations
Channel Operations
Collectives
Arithmetic Operations
Math Operations
Static Operations
Types
Examples and Tutorials
Examples and Tutorials
Overview
Running The Examples
Compile Diagnostics
Simple Matmul
Weight-Quantized Matmul
Static Multi-Core Scheduling
Tensor Memory Patterns
Math And Masking
Broadcasting
Static Evaluation
Causal Attention
Release Notes
Runtime
Runtime
Overview
Programming Guide
Programming Guide
Runtime API
Runtime API
Overview
Execution Model
Memory Model
Streams and Events
Multi-Device
Error Handling
Troubleshooting
Weight Conversion
Driver API
Driver API
Overview
Execution Model
Memory Model
Streams and Events
Error Handling and Reset
API Reference
API Reference
Overview
Python API
Python API
Overview
Device
Memory
Execution
Streams
Events
TLB
VMM
Sampling
Weight Converter
Types
C++ API
C++ API
Overview
Runtime API
Runtime API
Device
Memory
Execution
Streams
Events
TLB
VMM
Sampling
Types
Callback Params
Weight Converter
Driver API
Driver API
Overview
Device Handle
Memory
Execution
Streams & Events
TLB
Types
Environment Variables
Error Codes
Support Matrix
Release Notes
Torch
Torch
Overview
Programming Guide
Programming Guide
Overview
Device Semantics
Operations
Distributed
Paged KV Cache
Migrating from CUDA
Error Handling
Troubleshooting
API Reference
API Reference
Overview
Device API
Custom Operations
VMM API
Supported Operations
Environment Variables
Support Matrix
Release Notes
Torch Legato
Driver
Driver
Overview
Operations Guide
Operations Guide
Overview
Features
Features
Telemetry
Hang Detection and Watchdog
FATAL and Recovery
Firmware Update
Multiple Devices
Tools
Tools
hyperaccel-smi
ha-hotreset
ha-fw-update
ha-led
ha-wipe
Reference
Reference
sysfs Nodes
Kernel Log Messages
Troubleshooting
Kernel Module Parameters
Kernel Module Parameters
Configuration and Usage
Parameter Reference
Examples and Tutorials
Examples and Tutorials
Overview
Collecting Telemetry with Prometheus
Building a Card Health Check
Collecting Diagnostics
Recovering an Unresponsive Card
Updating Firmware Across Cards
Release Notes
Profiler
Profiler
Overview
Programming Guide
Programming Guide
Overview
Collecting Traces
Device Tracing
Reading a Trace
Annotations
PyTorch Profiler Integration
Troubleshooting
API Reference
API Reference
Overview
haprof CLI
Python API
Python API
Overview
Annotations
Kineto
Counter Reference
Environment Variables
Support Matrix
Release Notes
CCL
CCL
Overview
Programming Guide
Programming Guide
Overview
Communicators
Collectives
Groups
Point-to-Point
Error Handling
Troubleshooting
Multi-Node
API Reference
API Reference
Overview
Python API
C++ API
C++ API
Overview
Communicators
Collectives
Point-to-Point
Groups
Types
Environment Variables
Error Codes
Support Matrix
Release Notes
Kernel
Kernel
Overview
Programming Guide
Programming Guide
First Kernel
Matrix Multiplication
Multi-Core Addition
Model Serving
Model Serving
vLLM
vLLM
vLLM HyperAccel
Tutorials
Tutorials
Qwen3 8B
Qwen3 30B-A3B (MoE)
Gemma-4 31B
EXAONE-4.5 33B
Llama-2 7B
Features
Features
Supported Features
Context and KV Cache
LPU Sampling
Using Multiple LPUs
Quantization
OpenAI-Compatible Server
Profiling
Configuration Guide
Troubleshooting
Cloud Native
Cloud Native
Overview
Containers
Kubernetes
Releases
OpenAI 호환 API
¶
이 페이지는 옮겨졌습니다.
OpenAI 호환 서버
를 참고하세요.
Back to top