Model Information

Display Name: HiVision

API Model ID: hinow/hivision

Category: Image To Text

Description: HiVision is Hinow's multimodal vision model with 124B parameters, delivering frontier-level image understanding for documents, charts, and natural images. Built on open-weight architecture. **Key Features:** - 128K token context window - Supports up to 30 high-resolution images - 124B multimodal decoder + 1B vision encoder - Handles images of any aspect ratio - Strong text performance maintained - Open-weight architecture **Capabilities:** - Image understanding and analysis - Document OCR and extraction - Chart and graph interpretation - Visual question answering - Multi-image reasoning - Technical diagram analysis **Best For:** - Document processing pipelines - Visual data extraction - Complex image analysis - Multi-image workflows - Enterprise vision applications **Technical Specs:** - Parameters: 124B (123B decoder + 1B encoder) - Architecture: Multimodal Transformer (Open-Weight) - Vision: Any resolution/aspect ratio - License: Hinow Enterprise

Context Window: 131,072 tokens

Max Output: 16,384 tokens

How to Use This Model

To use HiVision via the HInow.ai API, use the model ID: hinow/hivision

API Request Example (Chat/Text)

POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "hinow/hivision",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}
              

API Request Example (Image Generation)

POST https://api.hinow.ai/v1/images/generations
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "hinow/hivision",
  "prompt": "Your image description here"
}
              

Pricing

• input: $3.00
• output: $9.00

Available Parameters

• temperature: Controls randomness (Options: 0, 0.3, 0.5, 0.7, 1.0)
• top_p: Nucleus sampling (Options: 0.1, 0.5, 0.7, 0.9, 1.0)
• max_tokens: Max tokens (Options: 1024, 2048, 4096, 8192, 16384)

Quick Reference

To use this model, set: "model": "hinow/hivision"

Featured: Yes

Documentation: https://hinow.ai/models/hinow/hivision

API Endpoint: https://api.hinow.ai/v1