LLM 模型排行

Tool-native dataset generation skill for Codex, Claude Code, and Antigravity with SFT/DPO pipelines, verification, deduplication, and flexible export.

Implementation of build an LLM from scratch by Sebastian Raschka.

Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.

This tool provides a web interface to define domain-specific languages and uses large language models (LLM) to interact with those grammars via chat.

Redis Vector Library (RedisVL) -- the AI-native Java client for Redis.

Reference Java & Spring implementation of the Open Ingestion Standard (OIS). It provides a secure, decoupled, and vendor-neutral enterprise data integration platform leveraging modern Java 25 features, Spring Boot, Spring AI and vector search infrastructure to orchestrate data flows

Build Fullstack GenAI apps and AI Agents

Unraid CA template and Docker build for Mem0 (OpenMemory). This mega-container utilizes s6-overlay to seamlessly bundle the pre-compiled Qdrant Vector Database, the Python FastAPI/MCP server, and the Next.js Dashboard UI into a single, click-and-play local AI memory layer. Built for homelabs, privacy, and true agentic AI memory retention.

Code from the ODSC Agentic Graph RAG workshop combining vector, FTS & graph retrieval for RAG. Includes observability and guardrails for evaluating outputs.

Vectorify package for Laravel

Local AI load balancer for Ollama and MLX fleets. Auto-discovery, smart routing, OpenAI-compatible API, zero config. Perfect for Mac Minis & Studios.

A project combining roguelike with LLMs, RAG, Text2Speech, and Speech2Text

Desktop AI agent for Windows — an architecture that expands what any model can do. Image/video OCR, offline speech transcription, an agent-driven browser, real .xlsx/.pptx/.docx output, MCP, and a delegated team of agents. Runs on your machine, with any provider.

Multimodal AI Assistant is an advanced chatbot designed for local interactions with LLMs, files, and multimodal functionalities. Built on the Chainlit framework, it integrates Langchain and ChatOllama for hosting models locally, with cloud storage and monitoring via LiteralAI.

A curated list of tools, frameworks, and resources for running AI locally — inference, RAG, orchestration, monitoring, and more.

Layered prefill changes the scheduling axis from tokens to layers and removes redundant MoE weight reloads while keeping decode stall free. The result is lower TTFT, lower end-to-end latency, and lower energy per token without hurting TBT stability.

Network-faithful simulation of LLM serving and training deployments

A Scheduler for Batched LLM Inference

A comprehensive GUI wrapper application for Apple's Foundation Models Adapter Training Toolkit, providing an intuitive interface for training LoRA adapters for Apple's on-device foundation models.

Deferred Continuous Batching in Resource-Efficient Large Language Model Serving (EuroMLSys 2024)