Open-source agentic IDE that orchestrates multiple AI coding agents in parallel with support for frontier, open-weight, and locally hosted models.
stagewise is an open-source (AGPLv3) agentic IDE and orchestrator for software engineers that runs multiple coding agents in parallel, each independently implementing features or fixing bugs. It supports frontier models (Claude, GPT, Gemini), open-weight models (DeepSeek, Qwen, MiniMax), and locally hosted models (Ollama, vLLM, Llama.cpp) through a single interface with no lock-in. The agent maximizes cache-hit rates and uses compact state deltas and automatic context compression to keep long-running tasks token-efficient and low-cost. Users can bring their own API keys or use stagewise Cloud Inference with no external subscriptions required. stagewise is a product of stagewise.
For people
For agents
Nothing listed yet.