llm

➡️ Moved to Hivekeep: github.com/MarlBurroW/hivekeep — self-hosted autonomous AI agents. Your AI team. At home.

Self-hosted chat client for Ollama, OpenAI, and Anthropic. Web UI and terminal CLI, with web search and a browser-automation agent.

Pure RL to post-train base models for social reasoning capabilities. Lightweight replication of DeepSeek-R1-Zero with Social IQa dataset.

🤫 Token-lean sessions at the harness level. An easy to grasp output style, output-shrinking hooks, and log compression cut both input and output tokens.

Stop agents from designing sloppy UI. Inspired by Impeccable, but without the plugin machinery. Just one SKILL.md file.

A Claude Code skill that turns vague requirements into production-grade specs — by interviewing you like a senior architect would.

End-to-end pipeline for seeing how LLMs actually process your prompts. Capture attention across every layer, render heatmaps and cooking curves, compare variants with evidence — not vibes.

Ask question over your Notion Database! A naive Retrieval-Augmented Generation (RAG) pipeline backed by Langchain and Streamlit

Leveraging LLMs for modernization through intelligent chunking, iterative prompting and reflection, and retrieval augmented generation (RAG).

Learning Azure AI with APIM, Semantic Kernel and LangChain.

JavaChain 用于快速搭建LLM应用，参考了 LangChain 的架构设计，基于Java8实现。

A tiny language model that talks like a house cat named Miso.

A ridiculously fast Python BPE (Byte Pair Encoder) implementation written in Rust

One free AI endpoint, every free tier behind it. Local OpenAI-compatible gateway routing across OpenRouter, Groq, NVIDIA NIM, Cloudflare Workers AI, and HuggingFace with automatic failover.

A high-performance Llama3 implementation using Micronaut and GraalVM Native Image

Containerfiles including AI, game servers, bootc and even a rickroll.

SeaTunnel is a multimodal, high-performance, distributed, massive data integration tool.

🚀 LLM-I: Transform LLMs into natural interleaved multimodal creators! ✨ Tool-use framework supporting image search, generation, code execution & editing

Native Windows vLLM 0.27.1 wheels: Python 3.13, PyTorch 2.13 + CUDA 13.0, SM 7.5-12.0 for RTX 20/30/40/50, OpenAI-compatible serving, FlashAttention/Rust, 10 KV formats, Multi-TurboQuant, and experimental CPU/RAM/NVMe prompt-KV offload - no WSL or Docker.

You put an agent into an Elpis, and it becomes Elpis; Be Elpis my friend.