Showing 100 skill frameworks sorted by GitHub stars — all open-source with proven community adoption.
State-of-the-art ML models for NLP, vision and audio
Framework for building LLM-powered applications
Meta's promptable image segmentation foundation model
High-throughput LLM serving with PagedAttention
Data framework for LLM applications over custom data
Unified fine-tuning framework for 100+ LLMs with WebUI
Microsoft utility to convert files and documents to Markdown
Microsoft's deep learning optimization library for scale
Build web demos and UIs for ML models in Python
Unified framework for scaling AI and Python applications
Facebook's library for efficient similarity search and clustering
Open-source vector database for scalable similarity search
Industrial-strength natural language processing library
Deep learning framework to train, deploy and ship AI products
HuggingFace library for diffusion model development
OpenAI's contrastive language-image pretraining model
Microsoft SDK integrating LLMs into applications
Official OpenAI Python client library
2-5x faster LLM fine-tuning with 70% less memory
Memory layer for AI agents and assistants
High-performance vector similarity search engine
Multi-type data labeling tool for ML training data
HuggingFace library for easy ML dataset loading and sharing
Programming—not prompting—language models
Efficient control and templating for language models
Platform for ML lifecycle: tracking, registry, deployment
Microsoft's graph-based RAG for complex reasoning over text
End-to-end NLP framework for search and QA systems
Apple's array framework for ML on Apple Silicon
Parameter-efficient fine-tuning methods including LoRA
Topic modelling and document similarity library
Open-source AI-native vector database
Multilingual sentence, paragraph and image embeddings
Unified API for 100+ LLMs with OpenAI format
Cross-platform ML inferencing accelerator by Microsoft
Open-source vector similarity search for PostgreSQL
ML experiments and data version control system
Approximate nearest neighbors library by Spotify
Natural Language Toolkit - classic Python NLP library
IBM's document parsing and understanding library
Integration platform for AI agents with 250+ tools
Chat with your data using natural language via LLMs
SDK for integrating 250+ tools into AI agents and LLMs
Open-source vector database with hybrid semantic search
Tensor library for machine learning on edge devices
Fast BPE tokenizer used by OpenAI models
Train LLMs with RLHF, PPO, DPO and reward modeling
Pre-train, finetune, and deploy 20+ LLMs on your hardware
Open-source implementation of CLIP vision-language models
Structured text generation for language models
Build and evaluate LLM-based AI flows
Simple and fast RAG system with knowledge graph support
Extremely fast tokenizers for modern NLP
Production LLM serving toolkit by HuggingFace
Run LLMs in production with BentoML
Structured outputs for LLMs using Pydantic
All-in-one open-source embeddings database
Open-set object detection with language grounding
Retrieval and embedding models including BGE series
Streamlined tool for easily fine-tuning AI models
Training and inference PyTorch at scale with minimal code changes
Pre-processing library for unstructured data (PDFs, docs, etc.)
Database for AI data with multimodal vector storage
Data-centric AI library for finding and fixing dataset issues
Python bindings for llama.cpp with OpenAI-compatible API
Type-safe AI agent framework built on Pydantic
Fast serving framework for large language and vision models
Structured RAG framework for enterprise LLM applications
Evaluation framework for RAG pipelines
Open-source LLM engineering platform for observability
Build, ship and run AI applications in the cloud
HuggingFace acceleration and optimization for inference
Build production-ready conversational AI applications
Secure sandboxed environments for AI code execution
Framework for evaluating language models on NLP tasks
8-bit and 4-bit quantization for LLM memory efficiency
AI toolkit for building natural language interfaces
Unit testing framework for LLM outputs and RAG pipelines
ML and LLM monitoring and evaluation platform
Test and evaluate LLM outputs and prompt quality
Tensor search engine for text and images
Open source AI search and recommendation engine
Add validation and correction guardrails to LLM outputs
NVIDIA toolkit for adding programmable guardrails to LLMs
Serverless vector database for AI applications
Easy GPTQ model quantization for LLM deployment
AI observability platform for LLM tracing and evaluation
PyTorch-native finetuning library for LLMs
Efficient LLM compression, deployment and serving toolkit
Python bindings for GGML/GGUF quantized models
Use ColBERT and late-interaction models in RAG pipelines
Query language and runtime for large language models
Official Python library for Anthropic's Claude API
Blazing fast inference for text embeddings
Dataclass for multimodal data representation in ML
Superfast AI decision-making and routing layer for LLMs
Evaluation and tracking for LLM-based applications
Embedded AI inference library for desktop and edge
Open-source tools for testing and experimenting with prompts
Security toolkit for LLMs to detect prompt injection and PII