Skip to content
View aivrar's full-sized avatar
75 followers · 24 following

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Starstruckx2

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: Starstruckx2

Block or report aivrar

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
aivrar/README.md

Hey, I'm aivrar

I love AI, I love coding and even better I love coding with AI. I will work for VRAM.

I build portable, zero-install AI tools and apps for Windows: extract a folder, double-click, and go. No Docker, no system Python, no admin rights. Most of my projects are built fast with heavy AI-assisted coding.

Support my work

I'm an independent maintainer with no company or funding. If these free, MIT-licensed tools save you time, you can sponsor me on GitHub. Sponsorship goes to GPU time, testing machines and the AI coding subscriptions that keep these projects maintained. Thank you!

Supporters: be the first. Supporters who opt in get their name listed here.

Featured projects

Project What it does
Portable Hermes Agent Portable AI agent desktop for Windows with 100 tools, a GUI, local models via LM Studio, TTS, music, ComfyUI and workflows. No install, no Docker.
LocalSoundsAPI Offline AI audio studio with a web UI and local REST API: XTTS, Fish Speech, Kokoro, Stable Audio, ACE-Step, voice cloning and Whisper.
World Media (Open Media Explorer) One portable EXE for internet radio, live TV, podcasts and public-domain films from 11 open sources. No account, no telemetry.
vLLM for Windows Native Windows builds of vLLM with FlashAttention, OpenAI-compatible serving and KV-cache compression. No WSL, no Docker.
ComfyUI Portable Installer One-click portable ComfyUI with multi-GPU instances, 101 pre-built models, curated custom nodes and SageAttention.
Portable TTS Server V2 Portable speech studio with 19 TTS engines plus Whisper, voice cloning, subtitles and an audio editor (WSL2).
Brave Origin for Windows Run the Linux-only Brave Origin Nightly browser on Windows in one portable window.
Multi-TurboQuant Unified KV-cache compression for LLM inference: 12 methods, routing, capacity simulation and multi-GPU planning.

More: AgentNate · n8n + Python Portable · Portable Music Server · AI Character Engine · Fliporium P2P · all repositories

Articles

The common thread

  • No setup headaches: no system Python, Git, Docker or CUDA toolkit to install
  • No admin rights: everything runs in user space
  • Portable: copy the folder anywhere and it still works
  • Offline-capable: works without internet after first setup
  • Clean removal: delete the folder and it's gone

Tech & tools

Python PyTorch CUDA Node.js TypeScript Go Windows HuggingFace

image

Popular repositories Loading

  1. portable-hermes-agent portable-hermes-agent Public

    Hermes Agent made portable desktop for Windows — 100 tools, GUI, local models via LM Studio, TTS, Music, ComfyUI, workflows, tool maker. No install. No Docker. No admin rights.

    Python 232 48

  2. vllm-windows-build vllm-windows-build Public

    Native Windows vLLM: 0.27.1 Latest; 0.29.0 Python 3.14/CUDA 13.2 release (CPU TorchAudio), plus Python 3.13/CUDA 13.0 prerelease. PyTorch 2.13, FlashAttention/Rust, OpenAI serving, 10 KV formats, M…

    Python 59 8

  3. brave-origin-for-windows brave-origin-for-windows Public

    Brave Origin Nightly on Windows — free, portable, no installer. The Linux-only minimalist browser running inside WSL2 with a dark tabbed UI, live logs, in-app terminal, and one-click updates.

    HTML 57 1

  4. multi-turboquant multi-turboquant Public

    Unified KV-cache compression for LLM inference: 12 Python-native methods, guarded add-on composition and routing, analytical capacity simulation, Godzilla KVarN/TriAttention with SM86/SM89 qualific…

    Python 29 8

  5. LocalSoundsAPI LocalSoundsAPI Public

    Portable offline AI audio studio with web UI & local API – XTTS, Fish Speech, Kokoro, Stable Audio, ACE-Step, voice cloning, music gen (no install)

    Python 26 3

  6. comfyui-portable-installer comfyui-portable-installer Public

    One-click portable ComfyUI installer for Windows. No Docker, no Python, no Git, no admin rights. Downloads everything automatically. Fully relocatable.

    Python 12 4