NJIR-NOW ” Sovereign AI Series

5 Proprietary AI Models. One Unified Ecosystem. Zero Compromises.
Forged by NJIR.AI Part of the NJIRLAH Global Ecosystem.

Overview

NJIR-NOW is the next-generation sovereign AI model series developed by NJIR.AI — engineered for maximum versatility across Edge Computing, Deep Reasoning, Professional Intelligence, Fact Verification, and Multimodal Vision.

Each model in the NJIR-NOW lineup is purpose-built for a specific domain, yet seamlessly interoperable within the NJIR.AI ecosystem. From ultra-lightweight nano models that run on edge devices, to massive 30B-parameter reasoning engines capable of complex multi-step problem solving — NJIR-NOW covers the full spectrum of enterprise AI needs.

Model Lineup

njir-now:1.5b-nanoEdge AI

ollama run njir/njir-now:1.5b-nano

💡 Nano is the lightest model in the NJIR-NOW series. Designed for deployment on resource-constrained environments where speed and efficiency are paramount. Perfect for chatbots, quick Q&A systems, and lightweight automation tasks.

🧠 njir-now:30b-cascade — Deep Reasoning & Thinking

ollama run njir/njir-now:30b-cascade

🧠 Cascade is the flagship reasoning engine. Built on a cascading neural architecture, it excels at breaking down complex problems into manageable steps. Features native tool-use and deep thinking capabilities for autonomous problem-solving workflows.

🏢 njir-now:22b-pro — Professional Grade Intelligence

ollama run njir/njir-now:22b-pro

🏢 Pro is the workhorse of the NJIR-NOW family. Optimized for professional environments where reliability, accuracy, and consistent output quality are non-negotiable. Ideal for business intelligence, document processing, and enterprise-scale content creation.

🛡️ njir-now:7b-sentinel — Fact Verification & Checking

ollama run njir/njir-now:7b-sentinel

🛡️ Sentinel is your truth guardian. Specifically fine-tuned for fact-checking and claim verification tasks. Use it to validate outputs from other LLMs, cross-reference retrieved documents in RAG pipelines, or build trust-critical applications where accuracy is everything.

👁️ njir-now:8b-vision — Multimodal Vision AI

ollama run njir/njir-now:8b-vision

👁️ Vision brings the power of sight to the NJIR-NOW ecosystem. Process images, analyze visual content, extract text from documents, and engage in rich multimodal conversations. The only model in the series with native image understanding capabilities.

🚀 Quick Start

Install any NJIR-NOW model

# Choose your model based on your needs:
ollama pull njir/njir-now:1.5b-nano      # Lightweight edge AI
ollama pull njir/njir-now:7b-sentinel    # Fact checking
ollama pull njir/njir-now:8b-vision      # Vision + Text
ollama pull njir/njir-now:22b-pro        # Professional grade
ollama pull njir/njir-now:30b-cascade    # Deep reasoning

Run via CLI

ollama run njir/njir-now:22b-pro

Run via Python

from ollama import chat

response = chat(
    model='njir/njir-now:22b-pro',
    messages=[{'role': 'user', 'content': 'Analyze this quarterly report...'}]
)
print(response.message.content)

Run via cURL

curl http://localhost:11434/api/chat \
  -d '{
    "model": "njir/njir-now:30b-cascade",
    "messages": [{"role": "user", "content": "Solve this step by step..."}]
  }'

Run via JavaScript

import ollama from 'ollama'

const response = await ollama.chat({
    model: 'njir/njir-now:8b-vision',
    messages: [{ role: 'user', content: 'What do you see in this image?' }],
})
console.log(response.message.content)

🔗 Compatibility

NJIR-NOW models are natively compatible with:

📊 Local Performance Benchmarks

🏅 Speed Ranking (Tokens/Second)

🥇 njir-now:nano     ████████████████████████████████████████  145.9 tok/s
🥈 njir-now:vision   █████████████████                         61.9 tok/s
🥉 njir-now:sentinel ███████████████                           56.4 tok/s
👑 njir-now:pro      ███████                                   28.4 tok/s
👑 njir-now:cascade  █████                                     19.2 tok/s

Tested locally on standardized consumer hardware. Pro (22B) and Cascade (30B) require enterprise-tier RAM (32GB+) for optimal performance.

🏛️ About NJIR.AI

NJIR.AI is the sovereign foundation for enterprise intelligence, delivering unified, high-precision models for Vision, Logic, and Automation. Part of the NJIRLAH Global Ecosystem.

Our Model Families

• njir.gen-4 — 4th Gen Autonomous Swarm Intelligence
• njir.gen-3 — Third-Generation Sovereign AI
• njir.gen-2 — Next-Gen Sovereign AI Ecosystem (2.3M+ Pulls)
• njir.ai — The Original Sovereign AI Suite (1.1M+ Pulls)
• njir-now — Sovereign AI Series (You are here)

âš¡ Built Different. Built Sovereign. Built by NJIR.AI. âš¡
© 2026 NJIR.AI — All rights reserved.