Open source harness evals: which agent harness does tasks the best, for the least cost?
Repository Index
104,105 visible repositories in the index
2,137–2,160 of 2,242 · Most stars
DecisionPivot: a diagnostic benchmark for decision models in LLM systems. Contrast pairs that differ in one decisive detail, generators with independent solvers, scorer and runners.
- benchmark
- calibration
- decision-making
- evaluation
The repository host gives no description.
DiscoveryRSI: Recursively Learning to Discover Valuable Hypotheses
- agent-framework
- agentic-workflow
- aiforscience
- llm-agents
Wrap Gemini CLI as an OpenAI-compatible API service, allowing you to enjoy the free Gemini Pro model through API!
[ICML 2026] Let LLMs invent and evolve languages for efficient reasoning.
Investor’s Last Exam: Can AI Agents Do Equity Investment Without Seeing the Future?
GLM 5.3 Abliterated or Uncensored AI model — 35 jailbreak prompts, model guide and API. One 7API key also covers Claude, GPT, Gemini, Grok, DeepSeek and Kimi.
- abliterated
- abliterated-models
- anthropic-compatible
- claude-code
BIA Bob is a Jupyter+LLM-based assistant for interacting with image data and for working on Bio-image Analysis tasks.
JevStyle: local decision models for AI agents. Classify, route and score with a probability per option. systemone-compatible local server, MCP tools, Claude Code guard. Open weights, Apache-2.0.
- agent-skills
- calibration
- claude-code
- decision-model
BLA blinds your agent. LASIK gives it its sight back. Static de-blinding CLI + MCP server that undoes BLA (Blind LLM Agents) hardening so coding agents and REA can read the real code again. Never executes the target.
- ai-agents
- anti-bla
- bla
- claude-code
Curated Gemini Omni prompts, API guide, and video examples — video generation, editing, camera control, style transfer, text rendering, and multi-modal workflows.
- ai-prompts
- deepmind
- gemini
- gemini-omni
A browser bookmarklet for exporting conversations with ChatGPT as markdown files.
- bookmarklet
- browser-bookmarklets
- browser-tools
- chatgpt
Your personal AI workbench on desktop. Execute tools, install skills, connect MCP servers, build persistent memory — all models, one interface.
TranslatedPi.NET is a .NET 10 port of Pi. Pi is a minimal, extensible Agent harness (a TypeScript monorepo of about 171,000 lines). This project ports its core libraries one by one to net10.0, so that the .NET ecosystem can use the same set of capabilities.
- agent
- agentic-ai
- dotnet
SUGT - local AI gateway & skills console. Centralize LLM API keys, one OpenAI/Anthropic-compatible proxy for Claude Code, Codex & OpenCode. Homebrew & winget. Fully open source (Apache-2.0).
- ai
- ai-gateway
- anthropic
- claude-code
ReAlign models instantly, on demand. Per-request alignment for verified users and channels, or Always On. Use directly or with Hermes, OpenClaw and custom harnesses.
.NET 7 SDK for OpenAI with a Blazor Server playground
- openai
- openai-api
- sdk
- sdk-dotnet
DeepSeek Harness Desktop 插件,让 AI Agent 用自然语言构建、检查和维护可编辑的 SideFX Houdini 工程,支持程序化建模、HDA、场景诊断与渲染。 | A DeepSeek Harness Desktop plugin that lets AI agents build, inspect, and maintain editable SideFX Houdini projects with natural language, including procedural modeling, HDAs, scene diagnostics, and rendering.
- dsh-plugin
- houdini
- ai-agent
Modern desktop application (Rust + Tauri v2 + Svelte 5 + Candle (HF)) for communicating with AI models that runs completely locally on your computer. No subscriptions, no data sent to the internet — just you and your personal AI assistant
- ai-chat
- artificial-intelligence
- candle
- chatbot
Set up Claude Code once. Your whole company gets it. Self-hosted Claude Code and Codex with your company files, team skills, MCP and per-person folder permissions.
- agent-skills
- ai-workspace
- claude-code
- codex
Public code release for the paper "The Surprising Effectiveness of Deleting Weights in LLM Post-Training"
A cross-platform desktop and web app to translate and validate SRT subtitles using your existing LLM subscriptions (Gemini, ChatGPT, Claude, etc.) via CLIProxyAPI - no API keys needed.
Integrate ChatGPT-style LLMs into Neovim: ask questions, keep multi-file context, pull diagnostics / command output, and chat without leaving your editor. Highly configurable, async, lua-only plugin.