Havuz

Qwen3.8-Flash-Next (125B MoE) up to 1M context on one or two RTX cards (tuned on 3080 10gb). A measured, tuned and benchmarked fork of Str

Telemetria cifrada con TypeScript, agente Go, analisis Python y Ollama, Cloudflare Tunnel y lectura mTLS.

Black-box LLM distillation for any verifiable task, with text or image inputs: teacher API -> rejection sampling -> SFT -> eval -> GRPO

Full-pipeline ComfyUI node pack tailored for Krea 2 Turbo: NAG negative guidance, Variance diversity, custom schedulers, samplers & aestheti

A 6.87 GiB GGUF build of Qwen 3.8 27B — full 27B quality on 8 GB RAM. 96.7% HumanEval, 92.9% MMLU retention, no distillation.

This project is for learning: it shows how an agent loop works underneath, in a way more intuitive and visual than reading about it. Behind

🔥🔥 Accelerating decision inference with exact state forking

Free, open-source connector agent chatbot AI browser to many App via MCP.

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

Private, offline AI assistant for Android and Windows: run local LLMs (Gemma 4, Qwen, any GGUF) on your phone or PC with llama.cpp, CUDA and

Source code of Local Code, a private AI agent for Mac, iPad and iPhone. Runs DeepSeek 4.1 on device, with Python, Node.js and a shell built

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

A 4B decision model built for Korean.

Run stock llama.cpp with the stock Unsloth Qwen3.6-35B-A3B 4-bit on a 16 GB Mac.

A computer, compiled into the weights of a language model. An exact CPU inside Qwen3.5-4B, with Gated DeltaNet state as RAM.

Official repository for MemT-1.0

qwen2API çok dilli (i18n) fork'u • Multi-language (i18n) fork of qwen2API — WebUI artık Çince, İngilizce ve Türkçe / UI now in Chinese, Engl

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, ll

Qwen3.8-Flash-Next (180B MoE) on one 12-16 GB NVIDIA GPU, Linux or Windows. Own engine, own compression.

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

SpeakMore：口述成文，划词即译。开源、免费的语音输入与翻译工具，支持 Mac 与 Windows：语音输入、语音翻译、划词翻译。兼容 Gemini、OpenAI、通义千问、DeepSeek、Kimi、智谱 GLM、MiniMax 及任意 OpenAI 兼容服务，语音输入亦

Native AI assistant for macOS, built with SwiftUI and llama.cpp. Run local LLMs offline, work with PDFs, and create custom AI assistants. Op

Carefully human hand written llm jailbreaks for the latest models that are available on the market. GPT-6 Astra, GPT-6.1 SoL, Opus 4.7, Opus

Pure-Rust MoE inference engine for Qwen3.5/3.6 GGUF. CPU+GPU hybrid experts, NVRTC CUDA kernels, OpenAI-compatible server. 2x llama.cpp pref

Local AI video dubbing studio with voice cloning, speech translation, speaker diarization and audio separation. Edit, preview and export dub

ComfyUI nodes for Strata: text, structured prompts, vision and managed local inference with GPU handoff. Publisher: t8star.

AI-assisted text editor for Windows with local LLM support, text rewriting, style improvement and Unicode cleanup.

Give the agent "this" and he gets a real PC.

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 fi

The fastest inference engine for Qwen3.8-27B on the NVIDIA RTX 5090: up to 500 tokens/s for one agent and up to 2,000 tokens/s for many, con

PS5LM, jailbreak yapılmış PS5 üzerinde llama.cpp tabanlı GGUF modellerini çalıştırmak için hazırlanmış bir homebrew payload’ıdır.

Neden yükseliyor · Bir gün yaşında 27 yıldız almış; PS5 üzerinde yerel LLM çalıştırma fikri ve gösterilebilir demo dikkat çekici. Ancak son 28 günlük commit ve farklı yazar verisi ölçülmediğinden ilginin sürdürülebilirliği ve geliştirme temposu hakkında kanıt yok.

Local decision model with calibrated probabilities: send a state and yes/no, choice or score questions, get a probability for every option.

A structured, versioned archive of adversarial prompt engineering techniques against Large Language Models — for AI safety research, red-tea

Run Qwen3.8-Flash-Next on ONE RTX 3090 (24 GB) + 64 GB RAM: 128K context, up to 2,100 tok/s prefill, 43–51 tok/s decode. vLLM runtime with h

Instinct: single-forward-pass decision models (choice / true-false / score) returning a probability per candidate

An environment where open models work as agents: a first-level harness with routing, skills, memory and sandboxed tools, on an engine that d

🚀🚀 A multimodal System One decision model that gives calibrated answers to typed questions about screens, photos, video and text in one forw

Distil an expensive LLM API call on a narrow task into a small local model: capture traffic, curate, LoRA fine-tune with MLX on Apple Silico

Small enough to run on your Mac. Smart enough to ask. An open-source computer use agent for macOS, with local models (MLX) that see, decide

collabosm, kullanıcının Google Colab hesabında GPU başlatıp Qwen3.8 modellerini yerel uygulama ve OpenAI uyumlu API üzerinden kullanmasını sağlar.

Neden yükseliyor · 3 günlük bir repoda 87 yıldız dikkat çekici bir başlangıç; ancak yıldızların neden geldiği ve kullanıcıların uygulamayı gerçekten çalıştırıp çalıştırmadığı mevcut verilerden anlaşılamıyor. Son 28 günlük commit ve farklı yazar ölçümleri de yok.

A 4B decision model: calibrated probabilities for yes/no, choice and score questions over long inputs.

High-performance, high-precision, locally-fast-deployed Jev-class medical decision service on national medical exams — four readings, one ro

A minimal yet ready-to-use AI agent harness, built on best practices for frontier models.

OpenJev: open-source, local alternative to Jev. System-One decision models: typed questions in, calibrated probabilities out; general decisi

🎨 Agentic skill for Qwen-Image-2.1: Rewrites and optimizes text-to-image and multi-image editing prompts using official Alibaba specificatio

网页版登录态 → OpenAI 兼容本地反代：DeepSeek / Qwen / Doubao 免费额度驱动 ZCode。Web login → OpenAI-compatible local API.

Hemmingway-1: a 27B model that writes the way a person writes. Open weights, free for non-commercial use (CC BY-NC 4.0); commercial use by a

An inference framework customized for AI MAX 395, compatible with both Linux and Windows, and supports deploying only qwen3.8-flash-next. Th

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

Run SemIf (Jev-style semantic-if decisions) on a CPU — no GPU. Reads typed option probabilities straight from an open model in one forward p