Neural Inverse - Open Source ("Neural Inverse OSS")
June 12, 2026 · View on GitHub
The Repository
This repository ("Neural Inverse OSS") is where we (Neural Inverse) develop the Neural Inverse open-source edition together with the community. Not only do we work on code and issues here, but we also publish our roadmap and iteration plans. This source code is available to everyone under the Apache License 2.0.
NeuralInverse
NeuralInverse is an AI-native IDE built for the work most AI tools ignore — modernizing legacy systems, developing firmware, and migrating regulated codebases. It is a distribution of the Neural Inverse OSS repository with additional features released under a commercial license.
NeuralInverse combines the editing experience of VS Code with purpose-built tooling for embedded engineers, enterprise architects, and teams maintaining safety-critical or regulated software. It is an open-source Cursor alternative with AI chat, inline edit, autocomplete, and an autonomous coding agent — plus specialized modules for firmware and legacy migration that no other AI IDE offers. Bring your own LLM — 20 providers supported, API keys never leave your machine.
NeuralInverse is updated regularly with new features and bug fixes. You can download it for Windows, macOS, and Linux on NeuralInverse's website.
Install:
# macOS / Linux
curl -fsSL https://neuralinverse.com/sh | bash
# Windows (PowerShell)
irm https://neuralinverse.com/win | iex
Neural Inverse Cloud
Use Neural Inverse directly in your browser at cloud.neuralinverse.com with forever-free models (Llama 3.3 70B, DeepSeek R1, DeepSeek V3/V4, Mistral Large 3, Kimi K2.6, and more). No API keys required, no configuration needed.
Features
- AI Chat & Inline Edit (Ctrl+L, Ctrl+K) — multi-mode sidebar chat, inline diffs, autocomplete, Fast Apply
- Power Mode (Cmd+Alt+P) — autonomous coding agent with 22+ tools and concurrent sub-agents
- Forever-Free Models — Llama 3.3 70B, DeepSeek R1/V3/V4, Mistral Large 3, Kimi K2.6 via cloud.neuralinverse.com
- Bring Your Own LLM — 20 providers (cloud, local, gateway), per-feature model selection, zero lock-in
- Firmware & Embedded (Cmd+Alt+F) — 357 MCU variants, SVD register maps, 22
fw_*agent tools, serial monitor, MISRA/CERT-C compliance - Legacy Modernisation (Cmd+Alt+M) — 5-stage migration pipeline, 30+ source languages, 61 translation profiles, Knowledge Base, audit export
- Agent Manager (Cmd+Alt+A) — model management, deployments, agent orchestration
Contributing
There are many ways in which you can participate in this project, for example:
- Submit bugs and feature requests, and help us verify as they are checked in
- Review source code changes
- Add new LLM provider integrations
- Add new language support to the modernisation engine
- Add new MCU/platform support to the firmware module
If you are interested in fixing issues and contributing directly to the code base, please see the document How to Contribute, which covers:
- How to build and run from source
- The development workflow, including debugging and running tests
- Coding guidelines
- Submitting pull requests
Full contributor guides are available at neuralinverse.com/guides/contributing:
- Getting Started — dev setup and first contribution
- Architecture — module map, DI, how features connect
- Bring Your Own LLM — providers, per-feature model selection, adding providers
- AI Chat & Inline Edit — sidebar chat, Ctrl+K, autocomplete, Fast Apply
- Power Mode — autonomous agent, tools, sub-agents, configuration
- Firmware & Embedded — MCU database, SVD, serial monitor, agent tools
- Legacy Modernisation — 5-stage pipeline, adding languages and profiles
- Model Management — deployment registry, cloud provisioning, agent manager UI
Feedback
- Request a new feature
- File an issue
- Email: github@neuralinverse.com
- Website: neuralinverse.com
Related Projects
| Module | Path | What it does |
|---|---|---|
| AI Chat & Core | src/vs/workbench/contrib/void/ | Sidebar chat, inline edit, autocomplete, LLM routing, settings |
| Power Mode | src/vs/workbench/contrib/powerMode/ | Autonomous agent with tool calling |
| Agent Manager | src/vs/workbench/contrib/neuralInverse/ | Model management, deployments, agent orchestration |
| Firmware | src/vs/workbench/contrib/neuralInverseFirmware/ | MCU database, SVD register maps, serial monitor, fw_* tools |
| Modernisation | src/vs/workbench/contrib/neuralInverseModernisation/ | 5-stage legacy migration engine |
Development Container
This repository includes a Visual Studio Code Dev Containers / GitHub Codespaces development container.
- For Dev Containers, use the Dev Containers: Clone Repository in Container Volume... command which creates a Docker volume for better disk I/O on macOS and Windows.
- For Codespaces, install the GitHub Codespaces extension in VS Code and use the Codespaces: Create New Codespace command.
Docker / the Codespace should have at least 4 cores and 6 GB of RAM (8 GB recommended) to run a full build.
Building from Source
npm install
npm run watch # Terminal 1: watch TypeScript
npm run watchreact # Terminal 2: watch React UI
./scripts/code.sh # Terminal 3: launch dev instance (macOS/Linux)
.\scripts\code.bat # Terminal 3: launch dev instance (Windows)
See HOW_TO_CONTRIBUTE.md for full platform-specific setup instructions.
License
Copyright (c) Neural Inverse Inc. All rights reserved.
Licensed under the Apache License 2.0.
Neural Inverse OSS is built on VS Code by Microsoft, licensed under MIT.