quantization-and-model-compression

SKILLWorkflowcommunity
v0.0.0jpoindexterUnknownUpdated 14d agoSource →

Reference-grade guide to shrinking and speeding up LLMs without retraining from scratch — numeric formats (FP8/INT8/INT4), PTQ methods (GPTQ, AWQ, SmoothQuant, bitsandbytes NF4, GGUF k-quants), KV-cache quantization, speculative decoding (Medusa/EAGLE/n-gram), and distillation — with concrete number

Community-submitted skill. Not yet reviewed by the Forge team. Full prompt content may not be available.Request review →
1Repo stars
1Clients
1Formats
14d agoLast update
Skill
Authorjpoindexter
Version0.0.0
LicenseUnknown
CategoryWorkflow
Formatsskill.md
PromptNot published
Compatibility
Claude✓ Supported
Cursor
Copilot
ChatGPT
Gemini
About

Reference-grade guide to shrinking and speeding up LLMs without retraining from scratch — numeric formats (FP8/INT8/INT4), PTQ methods (GPTQ, AWQ, SmoothQuant, bitsandbytes NF4, GGUF k-quants), KV-cache quantization, speculative decoding (Medusa/EAGLE/n-gram), and distillation — with concrete numbers, when each fits, and the quality cliffs.

Keywords
skillclaude