gptq
Post-training 4-bit quantization for LLMs with minimal accuracy loss.
majiayu000
7 published skills indexed across 4 categories. Combined upstream popularity: ★ 2,796.
16
indexed
7
published
3.6
avg score / 5
★3k
upstream stars
Active in 4 categories
7 published skills
hqq-quantization
Half-Quadratic Quantization for LLMs without calibration data.
peft-fine-tuning
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods.
clip
OpenAI's model connecting vision and language.
serving-llms-vllm
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching.
verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL).
ux
创建卓越的用户体验设计，包含用户研究、信息架构、交互设计、可用性评估。当用户需要设计界面、优化用户体验、进行可用性分析、创建用户流程、设计交互原型，或提到「ux」「用户体验」「交互设计」「可用性」「用户研究」时使用。产出专业、有洞察力的 UX 设计方案，避免通用化设计。