ResQ/ RAiS:7b

35 5 months ago

RAiS Visual Assistant 7B is a bilingual (English + Bahasa Melayu) multimodal AI model developed by ResQ based on llama 7B Vision It combines advanced visual understanding with natural-language reasoning, enabling text-and-image conversations

vision 7b 13b 
ollama run ResQ/RAiS:7b

Details

5 months ago

d0432dc34a4c · 4.7GB ·

model
llama
·
7.24B
·
Q4_0
projector
clip
·
312M
·
F16
license
Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US
template
{{ .Prompt }} ###
system
You are RaiS-va — a large language model developed by ResQ. You are intelligent, precise, and conc
license
ResQ Proprietary License — All Rights Reserved 2025
params
{ "repeat_penalty": 1.2, "stop": [ "###" ], "temperature": 0.6 }

Readme

🖼️ RAiS 7B, 13B — ResQ Vision & Language Assistant

RAiS 7B is a bilingual (English + Bahasa Melayu) multimodal AI model developed by ResQ.
It combines advanced visual understanding with natural-language reasoning, enabling text-and-image conversations, is a novel end-to-end trained large multimodal model that combines a vision encoder and Vicuna for general-purpose visual and language understanding — optimized for efficient local deployment and real-world AI-assistant applications.

⚙️ Model Overview

Property Details
Base Model llama 7B/13B + Vision Encoder
Model Type Multimodal (Text + Image)
Languages English, Bahasa Melayu
Input Resolution Up to 1344 × 1344 pixels
Temperature 0.7
Version RAiS v1.6
License ResQ Proprietary License — All Rights Reserved 2025

🚀 What’s New in RAiS v1.6

  • 🔍 Higher Resolution Input — Supports 672×672, 336×1344, 1344×336 for richer visual detail.
  • 🧠 Enhanced Visual Reasoning & OCR — Better scene analysis and text recognition from images.
  • 💬 Improved Visual Conversation — Handles a wider range of image-based queries naturally.
  • 🌏 Stronger World Knowledge & Logic — Upgraded instruction tuning for accurate, grounded answers.

🧩 Core Capabilities

  • Understands and responds in English and Bahasa Melayu.
  • Performs image captioning, description, and visual reasoning.
  • Handles OCR tasks and scene comprehension.
  • Provides concise, context-aware, and bilingual responses.
  • Ideal for AI assistants, multimodal agents, and edge deployments.

💻 Example CLI Usage

Pull and run the model locally:

”`bash ollama pull ResQ/RAiS:7b ollama run ResQ/RAiS:7b

🌐 Example API Usage

curl http://localhost:11434/api/generate -d ‘{ “model”: “RAiS:7b”, “prompt”: “Explain what ResQ represents in Malay.” }’

🌐 Image Prompt: curl http://localhost:11434/api/generate -d ‘{ “model”: “RAiS:7b”, “prompt”: “Describe what you see in this picture. Can i drive this car?”, “images”: [“iVBORw0KGgoAAAANSUhEUgAAAG0AAABmCAYAAADBPx+VAAA…”] }’

🔖 License ResQ Proprietary License — All Rights Reserved 2025 RAiS 7B/13B is intended for research and authorized commercial use under ResQ ’s AI development program.