Torad Labs ORG

QGRE Engine — Quality-Gated Reward Escalation for novel-domain GRPO training. Single-GPU, no Ray, step-level process rewards.

Engineering tenets enforced at write time, portable across coding-agent harnesses (codename; pre-release)

The Agent-Driven Architecture: an opinionated, platform-agnostic architecture for software where the agent — not the human — is the primary operator. A book + a seam-by-seam worked example.

Run Claude Code on your ChatGPT, Grok, or Kimi subscription — reasoning mirror, cache-warm compaction, and a fleet dashboard on loopback. Unofficial; use at your own risk.

Kotlin Multiplatform AI SDK agent, tool, streaming, and UI-message primitives

Organization profile and community health files for Torad Labs

An open-source AI agent that brings the power of Gemini directly into your terminal.

A high-throughput and memory-efficient inference and serving engine for LLMs

verl: Volcano Engine Reinforcement Learning for LLMs

User-friendly AI Interface (Supports Ollama, OpenAI API, ...)