Popular GitHub Repositories for Vram

Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Training released! Surpasses GPT-4o in ID persistence! Official ComfyUI workflow release! Only 4GB VRAM is enough to run!

HyperGen - Optimized inference and fine-tuning framework for diffusion (image & video) models. Up to 3x faster & 80% less VRAM.

This custom_node for ComfyUI adds one-click "Virtual VRAM" for any UNet and CLIP loader as well MultiGPU integration in WanVideoWrapper, managing the offload/Block Swap of layers to DRAM *or* VRAM to maximize the latent space of your card. Also includes nodes for directly loading entire components (UNet, CLIP, VAE) onto the device you choose

Play LLaMA2 (official / 中文版 / INT4 / llama2.cpp) Together! ONLY 3 STEPS! ( non GPU / 5GB vRAM / 8~14GB vRAM)

A complete full-stack voice generation solution with an intuitive user interface, project management, and advanced VRAM optimization features

Private voice keyboard, AI chat, images, webcam, recordings, voice control with >= 4 GiB of VRAM.

A repository for generating and training short audio samples with unconditional waveform diffusion on accessible consumer hardware (<2GB VRAM GPU)

A repository to run gpt-j-6b on low vram machines (4.2 gb minimum vram for 2000 token context, 3.5 gb for 1000 token context). Model loading takes 12gb free ram.

Run multiple resource-heavy Large Models (LM) on the same machine with limited amount of VRAM/other resources by exposing them on different ports and loading/unloading them on demand

A basic ui for running gpt neo 2.7B on low vram (3 gb Vram minimum)

Z-Image workflow with customizable image styles and GPU-friendly versions supporting various VRAM sizes.

Aims for memory-efficient training (24GB VRAM) on consumer GPUs. Optimizing language models through guidance tokens in reasoning chains, based on DeepSeekRL-Extended.

simple diffusers based implementation of Hunyuan-DiT, in Forge webUI for Stable Diffusion. Works with 8GB VRAM.

Stable Diffusion 3 in Forge webui, with controlNet, diffusers based, works with 8GB VRAM

[NVIDIA ONLY] Stable Diffusion WebUI Forge supporting Flux, Qwen, wan, nunchaku and more in a lightweight WebUI. (Minimum Requirements 6GB VRAM / 16GB RAM, Recommended Requirements 16GB VRAM / 48GB RAM)

Running Llama v2 with Llama.cpp in a 4GB VRAM GTX 1650.

[NVIDIA ONLY] AllTalk-TTS is a unified UI for E5-TTS, XTTS, Vite TTS, Piper TTS, Parler TTS and RVC. It supports a variety of advanced features, such as a settings page, low VRAM support, DeepSpeed, narrator, model finetuning, custom models, wav file maintenance. It can also be used with 3rd Party software via JSON calls.

[NVIDIA ONLY] Super Optimized Gradio UI for AI video creation for GPU poor machines (6GB+ VRAM). Supports Wan 2.1/2.2, Qwen, Hunyuan Video, LTX Video and Flux. https://github.com/deepbeepmeep/Wan2GP

VRAM calculator for Hugging Face models

An Agentic AI Research Playground for building AI Agents and Mult-Agent systems leveraging open-source models, libraries, and frameworks. Models used to run agents locally on Nvidia RTX A2000 GPU - 8GB VRAM include the quantized Ollama models: llama3.1:8b-instruct-q8_0, llama3.1:8b_q8_0, qwen2.5:7b-instruct-q8_0, and qwen2.5-coder:7b-instruct-q8_0.