Havuz

TokenSpeed, agentic workloads için LLM inference sunumunu hızlandırmayı ve istek yönetimini daha güvenli hale getirmeyi amaçlayan bir inference engine’dir.

Neden yükseliyor · 144 günde 2.178 yıldız ve son 28 günde 385 commit ile 20 farklı yazar, hem ilgi hem de yakın dönem geliştirme faaliyeti gösteriyor. README ayrıca PyTorch Ecosystem’e katılımı ve yeni model desteğini vurguluyor; performans iddiaları ise proje tarafından sunulan kıyaslamalara dayanıyor.

Project Maya - run GLM-5.3-Flash (321B MoE), DeepSeek-V4.1-Flash-1017 (748B MoE) and MiniMax-M3.1-Flash (428B MoE) privately on your own GPU

A native .NET LLM inference engine and agent runtime for GGUF models. TensorSharp provides a console application, a web-based chatbot interf

跨 AI 桌宠：跟着你用的 AI 换形象、跟着 AI 的工作状态做动作

Transformers for decision models: fine-tune, load and serve calibrated decsion-model on any Hugging Face model.

十五个 AI 客户端的内置提示词一键替换：双击即用、可自证、改前备份、一键还原。v8.26：一键破甲全程托管（自动结束进程+注入+自动拉起）、注入步骤实时提示、写后自检回滚、托盘常驻。免装 Python 单文件 exe，完全免费开源。

Open-source AI pair programming for desktop: a Mentor + Executor agent cross-check each other's code to catch AI hallucinations. Works with

Use 30+ AI models (DeepSeek, Kimi, GLM, Claude, GPT-5.5, Gemini, Grok) in GitHub Copilot Chat - free via BYOK. No Copilot Pro needed.

Per-window model switcher, subagent router, and automatic 429 failover for Claude Code — switch any window with /provider, delegate subagent

AI-native agent harness for coding workflows by python: multi-model LLM orchestration, stateful sessions, tool governance, traceable deliv

Lightweight Rust Agent only use 50MB RAM, but Claude Code Plugin compatible, Dynamic Workflow, Goal, Artifacts, Free Web Search, full featu

Run the pi coding agent as a self-hosted service: jobs from cron, the CLI, or GitHub, GitLab, Forgejo and Azure DevOps events, each in a loc

A coding agent engine and CLI with cache-aware context management.

Uncensored AI API and browser chat: DeepSeek and GLM models, streaming, Python and TypeScript examples, pricing and setup guides.

Uncensored & abliterated AI models (GLM-5.3, DeepSeek V4, Qwen3) + Claude, GPT, Kimi, Gemini — one OpenAI/Anthropic-compatible API key. Docs

SSD-streaming inference engine for giant MoE models (Rust + CUDA). GLM 5.2 743B at 2 tok/s and Hy3 295B at 7 tok/s on two consumer 16GB GPUs

Your pocket agent. Local-first AI agents on iOS and Android — real workspaces, tool execution with approvals, and your choice of model (DSH

Multi-Agent AI Coding Assistant - 支持 DeepSeek / GLM / MiMo 等 12+ 国产大模型，多智能体协作编程

Your friendly, blazing-fast, self-hostable AI gateway

IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse

Ultimate LLM API Integration Cookbook 2026 for Cursor & AI Agents

One ComfyUI skill for every AI coding agent (Claude Code, Codex, Gemini CLI, Qwen Code): drives a local ComfyUI end to end. 75 model prompt

Convert CodeBuddy/WorkBuddy (Tencent coding assistant) subscription into an OpenAI-compatible API. Use CodeBuddy GLM-5.2/Kimi/DeepSeek model

Use GPT, DeepSeek, GLM, Gemini, Kimi and 400+ models in Claude Desktop's own model picker and in Claude Code, while staying signed in to you

Illusion-Agent: Where fantasy meets functionality — an AI agent platform for terminal, browser, any model, any OS.

Claude orchestrates and reviews; Sonnet, Codex, GLM, or Grok implements. A Claude Code plugin for the orchestrator/implementer split: /serve

GLM-5.2, a 744 billion parameter mixture of experts model, in a pure C inference engine: quantized to int4, experts streamed from disk, depl

ACP agent in TypeScript that uses the Z.AI / Zhipu AI GLM Coding Plan models (GLM-5.3, GLM-5 Turbo, GLM-4.7) as the reasoning core

GLM-5.3-Flash EXL3 (320B MoE) on 2x NVIDIA DGX Spark — production serving kit, 1M context, 97%+ multi-session prefix caching, DFlash2 spec d

把腾讯 WorkBuddy 账号变成 OpenAI 兼容 API 的多账号网关。请求头照官方桌面端逐项对齐，主备入口 + HTTPDNS 自动切换，国内 / 国际双域合池，企业号可用 Claude，签到与成长任务全自动，React 管理面板支持一键更新。原版 workbuddy2

hummingbird is a lightweight, zero dependency runtime for massive open source Mixture of Experts (MoE) language models. It unifies SSD, RAM,

24/7 AI-DJ radio in Go. Persistent ffmpeg→Icecast with live ducking, LLM banter, BYOK, editable skills. ~50MB RAM.

Free AI models in VS Code Copilot Chat / DSH via browser auth. No API keys required.

Inference engine for large language models on consumer GPUs, with deep optimization for older hardware

Hybrid GPU + CPU inference for mixture-of-experts models, written in Rust down to the CUDA kernels. A drop-in llama-server.

Use DeepSeek, GLM, Kimi, OpenRouter and more in Codex CLI and Claude Code: one local endpoint, API translation, failover. Single Rust binary

A lightweight, general-purpose harness for tool-using LLM agents, fair benchmark evaluation, harness baselines, and personal assistant workf

GPT-5 / Claude Opus 4.7 / Gemini 3.1 / Sora 2 / Suno / DeepSeek / Kimi - AI API integration cookbook for Cursor, Cline, Claude Code, ChatBox

Windows 版 Codex Desktop 简体中文语言包，一键安装汉化；支持 codex汉化、Codex 中文菜单、国产模型接入、DeepSeek/千问/智谱、多账号切换、跳过官方登录。

QA agent swarms. Test web apps autonomously with greater speed and accuracy. Visual navigation, real workflows, evidence-backed reports.

Cross platform GLM proxy for AutoClaw's Zhipu AI models. Provides OpenAI-compatible Chat Completions and Anthropic Messages APIs for Claude

Claude省钱版 — 效果基本一致的情况下可以节约Token消耗约90%，超级省Token，支持OpenAI及国产大模型(DeepSeek/GLM/通义千问)的AI编程助手。本项目基于ClaudeCode源码实现。

Ultra cost-effective terminal AI coding assistant with excellent token cache hit rate. Built in ~10K lines of Go, it uses DeepSeek by defaul

Turn a model gateway into a standard OpenAI-compatible API with a built-in web console. Zero deps (Node.js built-ins only). 把华为云果办 OfficeAce

LinkTokenView（联视未来）国产大模型 Token API 聚合平台，提供 OpenAI 兼容接口，统一调用 DeepSeek v4.1、Kimi、GLM、通义千问 等模型，支持 Token Plan 与 Coding Plan。支持Codex、Claude、Harne

Any Model for VS Code Copilot

让 Codex 白嫖 NVIDIA 免费顶级模型，一键接入 | Transparent proxy between Codex and NVIDIA NIM — use NVIDIA's free hosted top-tier models with Codex

ZCode（Z.ai/GLM）AI 编程助手 JetBrains 插件：多标签对话、流式输出、模型与权限切换、任务/子代理可视化，AI 可经 browser-use 驱动内嵌 JCEF 浏览器 | ZCode (Z.ai/GLM) AI coding assistant for

vLLM recipe: GLM-5.3-Flash EXL3 K2 on one DGX Spark GB10. Native MTP k=2. Measured tok/s.