Havuz

A high-performance framework for training LLMs, VLMs, diffusion, and embodied models on NVIDIA GPUs and Kunlun XPUs. Supports Pi0.5, GR00T

ComfyUI nodes I made for my own workflows: click or type to mask a thing, crop, rotate and pad by dragging handles, inpaint and outpaint tha

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna

一套功能完整、轻量稳定的 ComfyUI 实用节点集。初衷是为了平替一些断更插件的功能，会持续更新一些新开发的节点插件。有好的创意可以发Issues，我会力所能及想办法做出来。全中文界面，原生控件，不依赖第三方库。

MetaClaw, agent konuşmalarından öğrenme sinyalleri çıkarıp skill güncellemeleri ve isteğe bağlı LoRA tabanlı RL eğitimiyle agent’ı zaman içinde iyileştirmeyi amaçlayan bir yazılımdır.

Neden yükseliyor · 3492 yıldız ve README’de duyurulan teknik rapor ile birden fazla agent entegrasyonu dikkat çekmesini açıklayabilir. Ancak son 28 günde commit ve farklı yazar sayısı sıfır; bu nedenle mevcut ivme veya bakım temposu doğrulanamıyor. Teknik raporun HuggingFace Daily Papers sıralaması README’de proje tarafından bildiriliyor.

Remiqora, yerel AI müzik üretimini stem ayırma, MIDI transkripsiyonu, LoRA eğitimi ve multitrack DAW iş akışıyla tek arayüzde birleştirir.

Neden yükseliyor · Birden fazla yerel müzik üretim ve işleme aracını tek uygulamada toplaması ilgi çekici; 10 günlük yaşına rağmen 125 yıldız ve son 28 günde 59 commit ile erken görünürlük ve geliştirme hareketi var. Üç farklı yazar katkısı olumlu bir etkinlik sinyali olsa da bu kısa geçmiş, kalıcı benimsenme veya olgunluk kanıtı değil.

Self-hosted, one-tab workbench for the whole training lifecycle: build Character/Concept/Style datasets (generate, scrape, or triage a big d

Halo is an open-source framework built by White Circle for training large language and multimodal models

Single-GPU LLM fine-tuning preflight: memory estimates, runnable LoRA/QLoRA recipes, and local measurements.

Fizgig — LoRA & Fine-tune Studio for Klein 9B, Krea 2, MiniMax H3, Qwen Image 2.1, Anima & Z-Image Turbo and Ace-Step XL: train, fine-tune,

The open .char format for consistent characters. Face, body, wardrobe & voice in one file, consistent across every model.

Your AI intranet: network the computers you already own for inference and training.

🦀 Decoder-only LLM built from scratch in pure Rust using Candle — no Python, no PyTorch. Gated DeltaNet + sparse attention, fine-grained MoE

Autonomous AI agent for Instagram — generates reels, carousels, memes, quote cards and posts them on autopilot. Free-tier LLM router, commer

Asking for What Was Never Requested: horizontal and vertical proactivity in LLM agents. Need-graph metrics (no LLM judge) and Q&D, which tra

Companion code for the Manning book 'LLM Customization and Fine-Tuning.' Adapt open-weights LLMs end to end: prompting and RAG, LoRA/QLoRA,

Train JEV-style System 1 decision models on Qwen3.5 and Qwen3.8-27B: 24-slot decision head, KL distillation, calibration and serving. 在 Qwen

Native macOS app for local AI image generation and editing on Apple silicon: Z-Image Turbo, Qwen-Image-2.1, Krea 2 Turbo, Anima Turbo and MO

A unified framework for training, fine-tuning, and evaluating World Action Models

Full-featured DAW, DJ, and VJ app interoperable w/Ableton, Reaper, Resolume & more. Stable Audio 3, Magenta RT2, Suno API, Chimera track fus

⚡ Zero-Stall MoE Inference via Lookahead Prediction & Async DMA Prefetching. Optimized for SSD I/O with Hybrid MLA+Sliding Window Attention.

Speedrunning LoRA fine-tuning: frozen task, frozen hardware, public wall-clock leaderboard. modded-nanogpt for fine-tuning.

The Best TypeScript framework for fine-tuning

A counterfactual battery for instruction following in vision-language-action policies: change one word of the instruction, hold the scene, a

LoRA fine-tune Gemma 4 31B to speak caveman-mode natively. Style: github.com/JuliusBrussee/caveman

MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transfor

Build your own offline AI from any documents. Free. No coding. LoRA fine-tuning + RAG + GGUF export.

Distill teacher chains-of-thought into a LoRA adapter via a strict boxed-answer format contract + two-phase Train→Nudge (silver-medal NVIDIA

🌹 Rose: Range-Of-Slice Equilibration PyTorch optimizer. Stateless optimization through range-normalized gradient updates.

Scaler 2 Music Theory Tool 2026: Direct DAW Integration for Windows 10 and 11

Self-hosted decision model (selfjev-4b) with a Jev-compatible API: Python SDK, server, fine-tuning and RLCD.

让 ComfyUI 像 WebUI 一样写提示词：中文翻译、Tag 自动补全、LoRA 管理、区域提示词控制，支持 Anima 与 SDXL 工作流。

🏆Winning Project | ModelGate is a contract-aware AI control plane that ingests customer contracts, extracts SLA/privacy/routing constraints,

AI Image Generator is a powerful and user-friendly desktop application designed to help creators produce stunning, high-quality artwork, pho

Local-first desktop manager for AI image libraries

Local inference platform for K/IQ-quant GGUF models on Apple Silicon

Teknofest 2026 Doğal Dil Ajanları Yarışması 1. Senaryo kapsamında geliştirilen, kamu sektörü için resmi evrak analizi ve taslak üretimi yapa

TK Toolkit (formerly Anima Toolkit) — all-in-one Anima/SD toolkit for ComfyUI: LoRA manager (batch apply + Civitai metadata & downloads), Da

A full-featured TUI MeshCore client for your terminal

Turn any local language model into a fast decision API with typed, probabilistic answers

A native macOS lab for teaching tiny language models to think — build the architecture, train the weights, and watch a small LLM emerge from

Dohnuts builds small multimodal models for direct decisions. -> System One model

Qwen Image 2.1 for DeepSeek Harness: native image editing on Apple Silicon MLX/MFLUX or self-hosted CUDA, LoRA, ControlNet, inpainting, pres

One GPU. Full LLM workflow. Real benchmarks. No cloud required.

Symbio is a self fine tuning AI agent that gives power to users through seamless interaction with their machines. This flexible system adapt

Local-first desktop studio for ComfyUI, built for AI short-drama production - zero-dependency 10MB exe: creation, gallery, 600+ templates, m

Local inference you can prove. A single-binary GGUF engine in C: serves, scores, and trains LoRA through the quantized weights it deploys, b

Local-first, end-to-end encrypted agent-to-agent mesh protocol. Zero-config LAN + LoRa. Runs offline. No cloud. info@ironmesh.org

Ultra-fast TensorRT engine for ComfyUI. Features native Anima2B & Flux support with near-instant LoRA switching via NVIDIA Refit. Reforged f

Free & Open Source AI Agent — Self-Hosted, Self-Learning, No Vendor API. Qwen2.5-7B + LoRA. 35 tools. Built on Islamic Epistemology (IHOS).