Tag

OpenCodeInstruct

Provides 5 million instruction–response pairs for supervised fine-tuning of code LLMs, with inputs, outputs, unit tests, and automated LLM judgments. Uses hybrid automated/synthetic generation and is released under CC BY 4.0 for large-scale SFT workflows.

Dyad

Builds apps locally with AI while keeping code, model choices, and deployment paths under the user’s control as a desktop alternative to hosted app builders.

MCP Shrimp Task Manager

Turns natural-language requirements into a dependency-aware graph of atomic, testable dev tasks for AI coding agents. Adds cross-session memory and a plan-reflect loop that forces the agent to think through each step before writing code.

DroidRun

Lets LLM agents drive real Android and iOS devices from natural-language commands by turning each screen's accessibility tree into structured text the model reads directly, not just screenshots. LLM-agnostic; runs via CLI, Python, or Docker.

trackers

Clean-room, modular implementations of multi-object tracking algorithms — SORT, ByteTrack, OC-SORT, BoT-SORT, C-BIoU — behind one interface. Detector-agnostic: works with YOLO, DETR, or any bounding-box model via supervision.Detections.

LiteRT-LM

Run large-language and multimodal models locally on edge devices (Android, iOS, desktop, web, Raspberry Pi) with hardware acceleration, function-calling, and multi-language SDKs—designed for low-latency, privacy-sensitive on-device inference.

Gemini CLI

Brings Gemini models into the terminal as an agent that reads files, runs shell commands, and edits code in place. Includes Google Search grounding, MCP server support, and a free OAuth tier (60 req/min, 1,000 req/day) with a 1M-token context window.

Awesome Claude Code

A curated index of community resources for Claude Code — skills, hooks, slash commands, agent orchestrators, and plugins. Entries live in a source-of-truth CSV that generates the README; submissions are bot-checked, then manually vetted by the maintainer.

Newton

GPU-accelerated physics simulation engine for robotics and simulation research — built on NVIDIA Warp with MuJoCo Warp backend, offering differentiable simulation, OpenUSD support, and extensions for RL/embodied-AI workflows. ([github.com](https://github.com/newton-physics/newton))

Chatterbox TTS

Open-source TTS that clones a voice from a short reference clip across 23+ languages, with adjustable emotional intensity via exaggeration/cfg controls and a built-in Perth neural watermark on every output.

V-JEPA 2 (facebookresearch/vjepa2)

Provides PyTorch code, pretrained checkpoints, and evaluation tooling for V-JEPA 2 — a Meta FAIR family of self-supervised video encoders and an action-conditioned world model. Includes training recipes, HuggingFace checkpoints, evaluation probes, and robot post‑training artifacts.

Microsoft Agent Framework

Builds production-grade AI agents and multi-agent workflows in .NET and Python, with graph-based orchestration for sequential, concurrent, and handoff patterns. Unifies Microsoft's Semantic Kernel and AutoGen lineages, adding durable, checkpointed runs.