--- language: - en library_name: peft license: apache-2.0 tags: - lora - conversational - text-generation - peft - fine-tuned - mamba - ssm - neuralai - base-model model_id: Subject-Emu-5259/NeuralAI inference: false --- # 🧠 NeuralAI: The Generative AI Engine

NeuralAI β€” Your AI. On your hardware. In your browser.

GitHub HF Hub Mamba K1 Mamba K2 Live UI

--- ## πŸ“Š Repository Composition | Language | Percentage | | --- | --- | | Python | 71.1% | | HTML | 13.0% | | JavaScript | 12.4% | | CSS | 2.6% | | Shell | 0.4% | | Jupyter Notebook | 0.3% | | Jinja | 0.2% | **The High-Velocity AI Engine for Your Entire Vibe Stack** NeuralAI is the central intelligence engine developed by **De'Andrew Preston Harris**. Conceived and engineered as an owned AI platform, it spans fine-tuned transformer models, custom SSM base models, DPO alignment, and a production web UI β€” all designed for local-first, private AI computing. --- ## πŸ—οΈ Model Family ```mermaid graph TB subgraph "NeuralAI Model Family" direction TB K1["🧬 Mamba K1
130M Β· SFT 50 steps
First Owned Base"] K2["🧬 Mamba K2
793M Β· Q4_K_M GGUF
460MB Β· Production"] K3["πŸ”¬ Mamba K3
SFT 500-1000 steps
10K+ UltraChat
In Training"] K1 --> K2 --> K3 K2 --> PROD["πŸš€ Active Inference
LM Studio Β· 460MB RAM
neuralai-web-ui.zocomputer.io"] end style K1 fill:#4a90d9,color:#fff style K2 fill:#22c55e,color:#fff style K3 fill:#f59e0b,color:#000 style PROD fill:#10b981,color:#fff ``` ### Complete Lineup | Model | Architecture | Params | Training | Status | Location | |-------|-------------|--------|----------|--------|----------| | **Mamba K1** | Mamba SSM | 130M | SFT LoRA 500 steps on 1K UltraChat (intel format) | πŸ”„ Retraining for chat coherence | `Subject-Emu-5259/NeuralAI-Mamba-K1` | | **Mamba K2** | Mamba SSM | 793M | Base pretrained β€” SFT queued (Q4_K_M GGUF) | ⚠️ Base model only | `Subject-Emu-5259/NeuralAI-Mamba-K2` | | **Mamba K3** | Mamba SSM | 2.8B | Base pretrained β€” SFT queued | ⚠️ Base model only | local `models/mamba-k3-base/` | ### Why Mamba SSM | Property | Benefit | |----------|--------| | Complexity | \(O(n)\) linear β€” scales to long context efficiently | | Inference | Fast at any sequence length, not just short prompts | | Memory | K2 runs at 460MB (Q4_K_M GGUF) β€” fits on any device | | Ownership | NeuralAI trains and merges all release weights on top of public Mamba SSM bases β€” every released GGUF is a fully merged model, not a raw base | | Ecosystem | LM Studio, llama.cpp, Hugging Face β€” mature deployment options | --- ## 🌟 Vision & Manifesto NeuralAI doesn't just predict text; it *operates the work*. The core mission is to create a multimodal generative system that bridges the gap between raw idea and execution. By fusing autoregressive generation with adaptive agency, NeuralAI becomes more than a chatbot β€” it is a persistent, reasoning partner. Born from resilience and ambition in Memphis, Tennessee and West Memphis, Arkansas, NeuralAI represents a forward-thinking approach to personal, private AI computing. --- ## πŸ› οΈ Tech Stack & Architecture NeuralAI is built on a high-performance architecture that decouples the inference engine from the web interface, enabling lightweight cloud hosting with powerful local inference. ### Core Stack - **Production Model**: Mamba K2 793M Q4_K_M GGUF (460MB) via llama.cpp β€” active inference engine - **Model Family**: Mamba K1 (130M, SFT LoRA retry) β†’ Mamba K2 (793M base GGUF) β†’ Mamba K3 (2.8B base) - **Inference Engine**: llama.cpp server with a custom `neuralai-intel` chat format (vocabulary-friendly for GPT-NeoX / Mamba tokenizers) - **Vocal Identity**: Andrew (Warm/Multilingual) β€” Optional voice synthesis integration - **Web Interface**: Custom Flask UI served via Zo Computer at `neuralai-web-ui-deandrewharris.zocomputer.io` - **Tool Chain**: 10 live slash commands (/web, /fetch, /browse, /research, /img, /speak, /summarize, /translate, /news, /yt) + NLβ†’Tool Router ### Future Scale Path | Stage | Params | Goal | Status | |-------|--------|------|--------| | **Mamba K1 SFT v2** | 130M | Finish 500-step SFT with intel format + GGUF | πŸ”„ Active | | **Mamba K2 SFT** | 793M | SFT LoRA 500 steps β†’ merge β†’ Q4_K_M GGUF | πŸ“‹ Next | | **Mamba K3 SFT** | 2.8B | SFT LoRA 1000 steps β†’ merge β†’ Q4_K_M GGUF | πŸ“‹ Next | | **Mamba 2B/3B** | ~2-3B | Scaled SSM architecture, benchmarks | πŸ“‹ Planned | --- ## ✨ Key Features & Capabilities ### πŸ’¬ Multimodal Chat & Agentic Intelligence - **High-Velocity Text Inference**: Fast, local inference with deep context awareness - **Deep Reasoning Mode**: Integration of test-time compute and chain-of-thought reasoning - **Autonomous Agentic Workflows**: Agent-mode interaction with browser, terminal, and third-party apps - **Live S2S (Speech-to-Speech)**: Real-time voice interaction with integrated microphone interface - **Identity Vault & Memory**: Persistent user memory and rule constraints ### πŸ”§ Developer & Engineering Tools - **10 Web Tool Commands**: Search, fetch, browse, research, image gen, TTS, summarize, translate, news, YouTube - **NLβ†’Tool Router**: Natural language web requests auto-routed to the right tool - **Model Manager**: CLI switching between all registered models - **Benchmark Suite**: Perplexity, generation diversity, MMLU-style, reasoning tests --- ## πŸš€ Model Lineage ```mermaid timeline title NeuralAI Model Evolution : Custom 135M base SFT 2026 Q3 : Mamba K1 β€” First owned base : 130M SSM Β· Proof of Concept 2026 Q3 : Mamba K2 β€” Scaled base : 790M Q4_K_M Β· GGUF ready 2026 Q3 : Mamba K3 β€” Full SFT : 500-1000 steps Β· 10K+ samples 2026 Q4 : Mamba 2B/3B β€” Next scale targets : 2B SSM β†’ 3B Core Intelligence ``` --- ## πŸš€ Deployment ```bash # 1. Start the inference service cd NeuralAI supervisorctl -c /etc/zo/supervisord-user.conf restart neuralai-lmstudio # 2. Start the web UI service python3 services/webui_service.py ``` ### Mamba K2 (LM Studio / llama.cpp) ```bash # Download from HuggingFace huggingface-cli download Subject-Emu-5259/NeuralAI-Mamba-K2 \ mamba-790m-hf.Q4_K_M.gguf --local-dir ./models/ # Serve with the NeuralAI chat format python3 services/lmstudio_server.py \ --model models/mamba-790m-hf.Q4_K_M.gguf \ --chat_format neuralai-intel \ --port 1234 ``` ### Mamba K1 (Python) ```python from transformers import MambaForCausalLM, AutoTokenizer model = MambaForCausalLM.from_pretrained("Subject-Emu-5259/NeuralAI-Mamba-K1") tokenizer = AutoTokenizer.from_pretrained("Subject-Emu-5259/NeuralAI-Mamba-K1") ``` ### Containerized Deployments | Deployment | Dockerfile | Stack | Status | | --- | --- | --- | --- | | **Web Chat** | `services/start_lmstudio.sh` + Flask UI | llama.cpp + `neuralai-web-ui.zocomputer.io` | βœ… Live | --- ## 🌌 NeuralAI Ecosystem The standalone software implementation of the NeuralAI core is **NeuralLabs**: πŸ‘‰ [github.com/Subject-Emu-5259/NeuralLabs](https://github.com/Subject-Emu-5259/NeuralLabs) **Software Downloads**: Latest beta builds available at: πŸ‘‰ [zo.pub/deandrewharris/neurallabs-beta](https://zo.pub/deandrewharris/neurallabs-beta) --- ## πŸ“ˆ Current State & Active Goals - **Legacy DPO v17 / Air 135M / SmolLM2-360M**: Retired and removed from the repository - **Mamba K1**: First owned SSM base model β€” retraining with intel-format SFT for coherent chat (Colab/GPU) - **Mamba K2**: Base pretrained GGUF ready β€” awaiting SFT - **Mamba K3**: 2.8B base downloaded β€” awaiting SFT - **Last Maintenance**: August 1, 2026 (Mamba Era β€” fix chat format + retrain pipeline) --- ## πŸ‘€ Creator Built by **De'Andrew Preston Harris** ([@deandrewharris94](https://linkedin.com/in/deandrewharris94/)) with Google Gemini AI Studio/Colab collaboration. From Memphis, Tennessee. Raised in West Memphis, Arkansas. AI Software Engineering at Maestro College. --- # NeuralAI β†’ Hugging Face sync is live