nasone32/llama.cpp-rdna3-7900xtx-opt
Open the project to explore the idea and the code.
01 / MACHINE INTELLIGENCE
Models you can run. Agents you can inspect. Tools you can build on.
← All topicsOpen the project to explore the idea and the code.
2 skills, composable into workflows, which can then be built into more complex workflows
An awesome collection of Jev use cases, workflows, and agent skills.
Open Managed Agents — an open-source alternative to Claude Managed Agents and OpenAI Agents API. Deploy with Node.js or Docker, run on Cloudflare, or use hosted OpenMA. Choose your models, harnesses and sandboxes.
Official code for "Self-Distilled Agentic Reinforcement Learning"
Open-source AI CMO for Claude Code: marketing agent skills for SEO, content, email, ads, launches, CRO, AEO/GEO, and AI-search visibility.
The security-first skill manager for AI agents — every install runs a security scan. Manage skills & MCP servers across 87 agents. Zero-dependency CLI.
Validated GLM-5.3 Flash recipe for 2x NVIDIA RTX PRO 6000 Blackwell 96GB: 262K context, EXL3/TR3, adaptive MTP, tools, and vision.
Agent skills for building with TypeSafe's System One API
DeepSeek V4.1 Flash on 4x RTX PRO 6000 Blackwell: native weights, DSpark, vision, NVMe or locked RAM Engram offload.
Turn DeepSeek Harness into a self-hosted enterprise Agent platform with knowledge, deep research, content generation, vibe coding, browser automation, governance, and admin controls.
The open-source System One decision model. Sub-15ms, non-autoregressive, local drop-in alternative to TypeSafe Jev.
A fast coding-agent harness in Go. Tool-use loop, bubbletea TUI, provider-routable models with live catalog discovery, MCP support, background subagents. One binary, no runtime, built for open-source models.
让任意大模型可以操控电脑,多平台可用,附 3D 建模案例与实机动态图。.
Qwen's most powerful open-source image generation model
Set up and improve harness engineering (AGENTS.md, docs/, lint rules, eval systems, project-level prompt engineering) for AI-agent-friendly codebases. Triggers on: new/empty project setup for AI agents, AGENTS.md or CLAUDE.md creation, harness engineering questions, making agents work better on a codebase.
One-pass typed decisions with calibrated probabilities (System One style model), fine-tuned from Qwen3.5-2B
A general agent harness for long term autonomous work, continual learning, and self-improvement.
A Unified Framework for Predicting, Designing, and Scoring Biomolecular Interactions
Build calibrated AI classifiers from human feedback using Jev and GEPA.
Open-source AI front-office factory for agencies serving local businesses: website, booking, intake, CRM, and agents in one workspace. Self-hostable.
Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.
Indexed knowledge bases with command-line tools for agents.
Open the project to explore the idea and the code.