Starred repositories
338
results
for sponsorable starred repositories
Clear filter
O(N·K) multi-head attention for PyTorch — a sparse drop-in replacement for dense scaled-dot-product attention.
The open-source CapCut alternative
GaetanLepage / tensordict
Forked from pytorch/tensordictTensorDict is a pytorch dedicated tensor container.
kakarot-dev / deepmiro
Forked from 666ghj/MiroFishOpen-source swarm intelligence engine — simulate hundreds of AI agents to predict outcomes. Includes MCP server for Claude.
# TurboQuant v3 (INT4 + AWQ + Protected Channels + Low-Rank) This notebook demonstrates a **TurboQuant-like** quantization algorithm: - Group-wise INT4 quantization - Activation-aware scaling (AWQ-…
Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1
An adaptive continually learning hybrid RAG Agent for content, curation, and persona growth. It combines spaCy-based NLP, a persona graph, BM25 retrieval, a truth gate, confidence scoring, and loca…
An experimental pi extension that runs and manages qwen with llama.cpp
Local-first cross-platform desktop workspace for Claude Code / agents: multi-agent, Git worktrees, code diffs, skill marketplace, multi-model, Computer Use, task-aware desktop pets, with WeChat, Fe…
TokenSpeed is a speed-of-light LLM inference engine.
A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io
Unsloth is a local UI for training and running Kimi K3, Gemma 4, Qwen3.6, DeepSeek-V4, GLM and other models.
Implementation of Inverse Reinforcement Learning (IRL) algorithms in Python/Tensorflow. Deep MaxEnt, MaxEnt, LPIRL
Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.
Maximum Entropy and Maximum Causal Entropy Inverse Reinforcement Learning Implementation in Python
Harness Engineering 学习指南 — 从概念理解到独立实践的深度学习档案
Agent skill for Codex, Claude Code, and OpenClaw that translates entire books (PDF/DOCX/EPUB) into any language using parallel subagents.
Lossless Claw — LCM (Lossless Context Management) plugin for OpenClaw
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
A PyTorch native library for training speculative decoding models
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
Claude Code skill for AMap (高德地图) Web Service API