[On-Device AI] 00. Introduction: On-Device AI 경량화
![[On-Device AI] 00. Introduction: On-Device AI 경량화](/post/on-device-ai-compression/getting-started-on-device-ai-compression/wordcloud_hu_4a9d67a17361f941.webp)
Collections
7 pages
[On-Device AI] On-Device AI 경량화
Pruning, Quantization, Knowledge Distillation을 CNN부터 LLM까지 다루고, Speculative Decoding·FlashAttention 같은 추론 가속 기법까지 정리하는 7편 시리즈입니다.
[On-Device AI] 01. Pruning — 가지치기로 모델 슬림화하기
![[On-Device AI] 01. Pruning — 가지치기로 모델 슬림화하기](/post/on-device-ai-compression/model-pruning-fundamentals/wordcloud_hu_d18c7288b3ba5937.webp)
[On-Device AI] 02. Quantization — 비트 수를 줄이는 법
![[On-Device AI] 02. Quantization — 비트 수를 줄이는 법](/post/on-device-ai-compression/model-quantization-fundamentals/wordcloud_hu_dd08274ddb7d3bc8.webp)
[On-Device AI] 03. Knowledge Distillation — Teacher와 Student
![[On-Device AI] 03. Knowledge Distillation — Teacher와 Student](/post/on-device-ai-compression/knowledge-distillation-fundamentals/wordcloud_hu_a8121718f0c3cbce.webp)
[On-Device AI] 04. LLM을 위한 Pruning — OBS와 SparseGPT
![[On-Device AI] 04. LLM을 위한 Pruning — OBS와 SparseGPT](/post/on-device-ai-compression/llm-pruning-fundamentals/wordcloud_hu_6b07358375e6c9c7.webp)
[On-Device AI] 05. LLM Quantization — SmoothQuant와 AWQ
![[On-Device AI] 05. LLM Quantization — SmoothQuant와 AWQ](/post/on-device-ai-compression/llm-quantization-fundamentals/wordcloud_hu_5ec605c5b13041ba.webp)
[On-Device AI] 06. 추론 가속 — Speculative Decoding, FlashAttention
![[On-Device AI] 06. 추론 가속 — Speculative Decoding, FlashAttention](/post/on-device-ai-compression/efficient-transformer-inference/wordcloud_hu_a31adae53dbca07d.webp)