Build LLM-powered applications in Ruby
Repository Index
111,821 visible repositories in the index
649–672 of 1,590 · Most stars
14-stage Fusion Pipeline for LLM token compression — reversible compression, AST-aware code analysis, intelligent content routing. Zero LLM inference cost. MIT licensed.
- ai-agent-tools
- context-compression
- context-pruning
- context-window-optimization
Easy token price estimates for 400+ LLMs. TokenOps.
One‑click codebase “blast” for Large‑Language‑Model workflows.
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
- interpretability
- mechanistic-interpretability
- qwen
- steering
NVIDIA AI Blueprint for video search and summarization (VSS) is a GPU-accelerated reference architecture for building video analytics agents with real-time verified alerts, visual Q&A, and automated reporting. The VSS Blueprint uses vision language models (VLMs) such as NVIDIA Cosmos, LLMs such as NVIDIA Nemotron, RAG, and NVIDIA NIMs.
An open-source implementaion for fine-tuning Qwen-VL series by Alibaba Cloud.
- multimodal
- qwen2-vl
- vision-language
- vision-language-model
CORTEX RAG is an enterprise retrieval and knowledge assistant that helps teams find accurate answers from company data with citations, permission-aware retrieval, and fast deployment
Custom Diffusion: Multi-Concept Customization of Text-to-Image Diffusion (CVPR 2023)
- customization
- fine-tuning
- text-to-image-generation
- computer-vision
中文羊驼大模型三期项目 (Chinese Llama-3 LLMs) developed from Meta Llama 3
- alpaca
- large-language-models
- llama
- llama-2
Ergonomically designed, AI-powered homework solver. | 符合人体工程学设计、人工智能驱动的作业助手 | 平庸者的苦工到此为止,这是来自外星的效率补丁 (by Gemini)
- ai
- homework
- self-learning
- self-taught
Supercharged experience for multiple models such as ChatGPT, DALL-E and Stable Diffusion.
- chatgpt
- dall-e
- stable-diffusion
- openai
An Open-source RL System from ByteDance Seed and Tsinghua AIR
designed specifically for AI painting (GPT, Midjourney, Nano Banana, etc.). Help users quickly build, manage, and iterate complex prompts through a visual "fill-in-the-blank" interaction. 通过可视化的"填空"交互方式,帮助用户快速构建、管理和迭代复杂的 Prompt。
Benchmark for tracking model capability after release.
Messaging APIs for multi-platform
- messaging
- messenger
- line
- api-client
[CVPR 2025] Open-source, End-to-end, Vision-Language-Action model for GUI Agent & Computer Use.
- computer-use
- vision-language-action
- vision-language-model
- agent
Run any Llama 2 locally with gradio UI on GPU or CPU from anywhere (Linux/Windows/Mac). Use `llama2-wrapper` as your local llama2 backend for Generative Agents/Apps.
- llama-2
- llama2
- llm
- llm-inference
Running Docling as an API service
Implementing DeepSeek R1's GRPO algorithm from scratch
One persona, many coding agents. Replace or inject system prompts across codex, claude, opencode, grok, omp, hermes, pi, zcode - with a local web console.
Text-To-Speech, RAG, and LLMs. All local!
- embeddings
- llm
- local
- rag
Super easy library for BERT based NLP models
- bert
- fast-bert
- fastai
- transformers
BullshitBench measures whether AI models challenge nonsensical prompts instead of confidently answering them, created by Peter Gostev.