LLM 模型排行

Build LLM systems you actually control. A free, open engineering book + course — from tokenization to serving your own models. Mechanisms, trade-offs, and numbers, not prompt tips

Code for the paper "Aligning LLM Agents by Learning Latent Preference from User Edits".

A meta-framework for self-improving LLMs with transparent reasoning

I recently attended the Geekbang "Large Language Models Application Development Practice Camp", where I learned about the application development of Large Language Models. Below is a summary of the after-class exercises that I participated in the camp.

Convert Confluence MIME exports (.doc) to clean Markdown

Efficient AI English Learning: Read & Speak via Web | 英语学习，英语单词，通过 AI 学英语朗读，对话的高效 Web 应用

LLMtranslator translates and generates text in multiple languages.

Spring AI agent engineering project with Java, MCP, RAG, tool calling, multimodal workflows, and UI prototypes.

Your agent seeks what search can't find. A self-hosted perception MCP server that transcribes speech, reads behind logins, sees images and video frames, crosses languages, and remembers.

Local-first CLI for benchmarking LLMs on real hardware — quality, speed, reliability, and a real multi-turn agent loop.

Emulates Azure OpenAI API on your local machine using Ollama and open-source models.

Python-side witness stack for the official DeepSeek Harness (Node). Five `dsh doctor --node` probes each report something the Node runtime will not. `pip install deepseek-harness-cli` first

code for paper Query-Dependent Prompt Evaluation and Optimization with Offline Inverse Reinforcement Learning

Makes your prompts better both Locally & Online, UI & NO UI

Generative AI-based CyberSecurity-focused Prompt Dataset for Benchmarking Large Language Models

The first Open-Souce RAG-LLM tool to analyse macroeconomic data and forecasts 🦜🔗

Streamlit + Langchain + LlamaCPP + Mistral + Rag

智能体系统设计与实战：基于langchain与langgraph，Agent System Design and Practical Applications: Based on LangChain and LangGraph

First open-source TurboQuant KV cache compression for LLM inference. Drop-in for HuggingFace. pip install turboquant.

Simple, Composable, High-Performance, Safe and Web3 Friendly AI Agents and LazAI Gateway for Everyone