Eidolon Chat Free
A Free Local AI Assistant by Blacknode LTD
Eidolon Chat Free is a lightweight but fully functional local AI assistant designed for privacy-focused offline usage.
Built on top of Gemma 4 E4B, this edition provides a practical and accessible local AI experience with multimodal capabilities, voice interaction, persistent local memory, document reading, and optional internet search.
Released under the Apache 2.0 License.
No subscriptions.
No forced cloud dependency.
No telemetry-focused architecture.
Because software should work for the user. Radical concept in 2026.
Features
🧠 Powered by Gemma 4 E4B
Optimized for fast and efficient local inference on consumer hardware.
👁 Vision Support
Analyze and discuss images directly inside the chat.
Supported workflows include:
- image understanding
- visual Q&A
- screenshot interpretation
- scene analysis
- diagrams and illustrations
📄 Document Reading
Upload and analyze documents directly inside the chat interface.
Examples:
- summaries
- explanations
- contextual Q&A
- study assistance
- document exploration
🎙 Voice Interaction
Integrated voice support includes:
- speech-to-text
- text-to-speech
- conversational interaction
🌐 Internet Search
Optional integrated web search allows the assistant to retrieve online information while maintaining a local-first workflow.
💾 Persistent Local Memory
Conversation logs can be stored locally to maintain continuity between sessions.
No cloud synchronization required.
Open Distribution Philosophy
Eidolon Chat Free is intentionally released:
- without DRM
- without code obfuscation
- without forced online activation
- without hidden cloud processing
This repository is designed to support:
- experimentation
- learning
- customization
- community-driven development
Model Details
Model Description
Eidolon Chat Free is a local AI assistant application built around the Gemma 4 E4B model family and designed as part of the broader Eidolon ecosystem developed by Blacknode LTD.
The project focuses on delivering accessible local AI functionality with multimodal support while remaining lightweight enough for consumer hardware.
- Developed by: Blacknode LTD
- Shared by: Blacknode LTD
- Model type: Multimodal Local AI Assistant
- Base Model: Gemma 4 E4B
- Language(s): Multilingual
- License: Apache 2.0
- Architecture: Local multimodal assistant application
Intended Uses
Direct Use
Eidolon Chat Free is intended for:
- local AI conversations
- document analysis
- image understanding
- voice interaction
- offline AI experimentation
- educational usage
- privacy-focused workflows
Out-of-Scope Use
This project is NOT intended for:
- medical diagnosis
- legal advice
- autonomous critical decision-making
- high-risk industrial usage
- fully autonomous agent execution without supervision
Users remain responsible for verifying outputs generated by the system.
Bias, Risks and Limitations
Like all large language models, Gemma-based systems may:
- hallucinate information
- generate inaccurate answers
- reflect biases present in training data
- misunderstand context
- produce incomplete or misleading outputs
Multimodal interpretation may also produce inaccurate analysis of visual or textual inputs.
Internet search functionality may retrieve inaccurate or outdated information from external sources.
Human supervision is recommended for important decisions.
Machines are still extremely confident pattern predictors pretending not to be probability engines. Sometimes they are correct. Sometimes they are very creative.
Hardware Requirements
Minimum
- Modern CPU
- 16 GB RAM
- SSD storage recommended
Recommended
- Apple Silicon Mac
- NVIDIA GPU with 8 GB VRAM+
- AMD GPU with Vulkan support
- 24 GB RAM or more
Installation
Windows
Download the latest release and run the installer.
macOS
Download the notarized DMG package and move Eidolon Chat Free into Applications.
Linux
Extract the package and launch the included startup script.
Ecosystem
Eidolon Chat Free is part of the broader Eidolon ecosystem developed by Blacknode LTD.
Additional commercial modules include:
- advanced AI orchestration
- multimodal generation pipelines
- astronomy modules
- reasoning systems
- specialized AI assistants
- advanced memory systems
Links
License
Apache 2.0 License
This project is released under the Apache License 2.0 to encourage experimentation, transparency, and local AI accessibility.
Acknowledgements
This project is built using technologies from the open-source AI ecosystem, including:
- Gemma
- llama.cpp
- Whisper
- Piper
- Hugging Face ecosystem
Special thanks to the local AI community for proving that consumer hardware is still capable of amazing things.
Model Card Authors
Blacknode LTD
Contact
