Showing 30 of 5,013 · Category: LLM
Mingli AI｜真一 ZHENYI，全球首个开源 AI 算命命理大模型。以盲派做功技法为核心，融合旺衰、子平、调候；程序自动排盘，模型负责推演。权重与源码开放，支持在线体验与独立部署。
Fine-tune Qwen2.5-Coder-7B for free (QLoRA on Kaggle T4) and run a local coding agent that runs your code instead of guessing — 8 GB Mac, llama.cpp, LoRA weight analysis with charts
Think only when the tool call looks unsure: confidence-gated reasoning for small local models. Code, BFCL results and paper.
Chinese AI learning notes and a minimal RAG demo with Sentence-Transformers, BM25, OpenAI-compatible APIs, and retrieval evaluation.
Use a PackyCode (PackyAPI) API key in DeepSeek Harness: built-in multi-protocol model route, ordered API-key rotation, and balance display
Windows utility focused on one-click local AI setup, local AI runtime, and Windows AI setup for local AI workflows.
AI-native, mixed-reality, self-evolving open world engine. 11-layer architecture, 5 platforms (PC / Mobile / Web / AR / BCI-reserved).
Windows utility focused on out of memory recovery, vram error, and local AI crash for local AI workflows.
Crossroad A2A: agent-to-agent mail so ChatGPT, Claude, Gemini, Codex and Grok agents work as one team across machines and SaaS apps. Open source, v0.1 beta.
基于 ARTEX 的 AI 自主渗透测试系统，支持多 Agent 协作、资产探索与漏洞管理，提供本地部署面板。
为 DeepSeek Harness 设计的破限插件：卸掉宿主的提示词限制和审批拦截，让助手照你说的做，不半路拒绝、不反复请示。补丁逐条可选，随时还原原版。 A jailbreak plugin for DeepSeek Harness: it strips the host's prompt restrictions and approval gates so the assistant does what you asked instead of refusing or asking permission mid-task. Patches are individually selectable and revert cleanly.
Open-source Android assistant that replaces Gemini. Summon it by gesture, bring your own API key, and it reads the screen, taps through apps and runs scripts. No backend, no telemetry.
Open-source temporal memory engine for everyday AI applications.
A decision library for running local LLMs on multi-GPU machines without NVLink: parallelism, communication, quantization, speculative decoding, KV cache. Every conclusion carries its evidence grade and reproducible conditions.
Strata 0.2.3 (engine 0.1.48) for AMD gfx906 (MI50/MI60/Radeon VII): rocBLAS solution table, Qwen3.6/Ornith engine restored, DFlash2 build fix, web configurator, portable (no AVX-VNNI needed). Docker: xxdoman/strata-mi50:latest
In-memory graph database and Gremlin traversal engine in Rust: browser playground, CLI, MCP server for AI agents, durable store, Python bindings.
Cut your AI coding agent's token bill, locally: Claude Code, Codex, Cursor, Copilot. #1 on THOL, 39% cheaper on long sessions.
Unified feature-rich agent kernel as service, build your own agent with a stable API
tinygrad on Google's Coral Edge TPU: reverse-engineered ISA, our own code generator and USB driver. A whole transformer runs on the stick. Written by AI agents.
Semantic search, cited answers and literature reviews for your own Zotero library – vibe-coded with Claude
Self-hosted LLM gateway: vendor keys, rule-based routing, fallbacks, throttling and usage reporting over gRPC
ИИ-ассистент в вашей базе 1С: читает конфигурацию и документы, отвечает по вашим данным, заполняет и создает документы. Бесплатная внешняя обработка .epf в Releases.
Super Mario 64's movement and collision code on the Game Boy Advance, bit-exact with the original, plus a playable demo. Built with Claude Opus 5.5.
OpenCode Android — unofficial client for opencode. The real coding agent runs natively on your phone: no Termux, no PC, no root. One arm64 APK for Android 8+, with file editing, shell commands, and wireless ADB app control.