DEMO LABS · DEMO RESOURCE

PRISM MULTIMODAL 2

Multimodal model combining language and vision for mixed text-and-image workflows.

Use in terminal Open playground

Capabilities

Multimodal analysisVisionLanguageStructured output

Modalities

textvision

Performance characteristics

Modality breadth: Text + vision
Latency profile: Balanced
Best for: Mixed media
Context: 128K context

Privacy characteristics

This resource supports transient processing — requests can be processed without being treated as permanent user history, where the underlying infrastructure allows it.

Compute pricing (demo)

Input: 1.2 credits
Output: 2.4 credits
Unit: per 1M units (demo)

API identifier

demo/prism-multimodal-2

Example request

curl https://api.intellex.ai/v1/generate \
  -H "Authorization: Bearer $INTELLEX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "demo/prism-multimodal-2",
    "input": "Your request here"
  }'

Illustrative shape only — the Intellex API is not live in this demo.

Demo interface — no live execution behind this panel