The Digital Futurist logoSkills and Prompt Registry
type public-role-directoryrole local-ai-engineercategory ai-ml/llm-systemscurator Thomas Cherickalupdated 2026-11-15

local-ai-engineer: Public Skills (20)

Sorted by GitHub stars of the source repo, highest first.

# Skill Owner Repo Stars License What it does
1 foundation-models-on-device affaan-m affaan-m/ECC 302,998 MIT Apple FoundationModels framework for on-device LLM — text generation, guided generation with @Generable, tool calling, and snapshot...
2 local-llm-expert sickn33 sickn33/agentic-awesome-skills 49,710 MIT Master local LLM inference, model selection, VRAM optimization, and local deployment using Ollama, llama.cpp, vLLM, and LM Studio. Expert...
3 ollama-stack sickn33 sickn33/agentic-awesome-skills 49,710 MIT Run local LLM workloads with Ollama, Open WebUI, and GPU-aware tuning for private development environments.
4 quantized-export wshobson wshobson/agents 42,512 MIT Export a promoted fine-tuned model in the right deployment format — merged safetensors, LoRA-only, GGUF with imatrix, or FP8. Use after a...
5 inference-serving-llama-cpp davila7 davila7/claude-code-templates 33,617 MIT Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel...
6 multimodal-whisper davila7 davila7/claude-code-templates 33,617 MIT OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language...
7 optimization-awq davila7 davila7/claude-code-templates 33,617 MIT Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models...
8 optimization-bitsandbytes davila7 davila7/claude-code-templates 33,617 MIT Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger...
9 optimization-gguf davila7 davila7/claude-code-templates 33,617 MIT GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or...
10 optimization-gptq davila7 davila7/claude-code-templates 33,617 MIT Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you...
11 optimization-hqq davila7 davila7/claude-code-templates 33,617 MIT Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing...
12 prompt-engineering-outlines davila7 davila7/claude-code-templates 33,617 MIT Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers,...
13 awq-quantization Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models...
14 gguf-quantization Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or...
15 gptq Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you...
16 hqq-quantization Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing...
17 llama-cpp Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel...
18 outlines Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers,...
19 quantizing-models-bitsandbytes Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger...
20 whisper Orchestra-Research Orchestra-Research/AI-Research-SKILLs 14,542 MIT OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language...