Gigaviz Studio Complete Guide: AI Documents, Charts, Images & More in One Creative Suite | Gigaviz
Feb 19, 2026
Gigaviz Studio Complete Guide: AI Documents, Charts, Images & More in One Creative Suite
A complete guide to Gigaviz Studio — the AI-powered creative suite that generates documents, charts, images, video storyboards, music compositions, and dashboards from a single prompt. Token-gated, multi-tenant, and built for business teams.
Gigaviz Studio Complete Guide: AI Documents, Charts, Images & More in One Creative Suite
Creating business content shouldn't require six different tools, three browser tabs, and a graphic designer on speed-dial. That's why we built Gigaviz Studio — a unified creative suite where you describe what you need, and AI builds it for you. Documents, charts, images, dashboards — all generated from a single prompt, all inside your workspace.
This guide covers everything Studio can do today, how the token economy works, and what's coming next.
What is Gigaviz Studio?
Gigaviz Studio is an AI-powered creative suite built into the Gigaviz SaaS platform. It has three modules:
Office — AI document automation (reports, invoices, proposals)
Graph — Data visualization (charts, dashboards, AI images, video storyboards)
Tracks — Workflow automation and AI music composition
Each module is gated by workspace entitlements, so you only pay for what you use. Every generation consumes tokens from your workspace balance — no surprise bills, no per-API-call pricing.
The 6 content types you can generate
1. AI Document Generation (Office)
What it does: Describe what you need — "Create a Q4 sales report for the Southeast Asia market" — and Studio generates a structured document with title, sections, headings, body text, and an executive summary.
AI Model: GPT-4o-mini Token Cost: 20 tokens per generation Output: Structured JSON document with title, multiple sections (heading + body), and a summary paragraph.
Use cases:
Monthly client reports from CRM data
Sales proposals with product descriptions
Internal memos and meeting summaries
Invoice drafts with line items
The document is stored as structured JSON, which means you can edit individual sections, re-order them, or regenerate specific parts without starting over.
2. AI Image Generation (Graph)
What it does: Type a prompt like "Modern tech startup logo with gradient blue and purple colors" and Studio generates a real image using DALL-E 3.
AI Model: DALL-E 3 Token Cost: 40 tokens per generation Output: High-quality image saved to your workspace storage (persisted permanently — not a temporary URL).
10 style presets:
Photo Realistic, Illustration, 3D Render, Watercolor
Pixel Art, Abstract, Flat Design, Anime
Logo, Icon
The style preset modifies the prompt sent to DALL-E 3, so "Logo" prepends instructions for clean vector-style output while "Photo Realistic" emphasizes photographic quality.
Why we persist to Supabase Storage: DALL-E 3 returns temporary URLs that expire. We download the generated image and re-upload it to your workspace's Supabase Storage bucket, so it's available forever.
3. AI Chart Generation (Graph)
What it does: Describe your data — "Monthly revenue chart for 2025 showing growth trend" — and Studio generates a complete chart configuration with sample data, labels, and visual settings.
AI Model: GPT-4o-mini Token Cost: 40 tokens per generation Output: Chart type recommendation + labels + datasets + configuration, rendered live by our ChartRenderer component.
Supported chart types:
Bar, Line, Pie, Area, Scatter, Radar
Funnel (for conversion analysis)
Heatmap (for correlation matrices)
Charts are rendered using Recharts with a custom dark theme. You can export any chart as PNG (2x retina) or SVG using the built-in export button.
4. AI Dashboard Generation (Graph)
What it does: Describe a dashboard — "Executive KPI dashboard for e-commerce store" — and Studio generates a responsive widget layout with stat cards, chart widgets, text blocks, and data tables.
AI Model: GPT-4o-mini Token Cost: 40 tokens per generation Output: Widget grid JSON rendered by the DashboardRenderer component.
Widget types:
Stat cards (metric + label + trend indicator)
Chart widgets (embedded chart configurations)
Text blocks (markdown-compatible)
Data tables (headers + rows)
Dashboards can be toggled between public and private. Public dashboards are shareable with stakeholders via URL — no login required.
5. AI Video Storyboard (Graph)
What it does: Describe a video concept — "60-second product demo for our WhatsApp CRM tool" — and Studio generates a storyboard with scene descriptions, narration script, and music suggestions.
AI Model: GPT-4o-mini Token Cost: 40 tokens per generation Output: Scene-by-scene breakdown with visual descriptions, timing, narration text, and a music/SFX recommendation.
8 video styles: Marketing, Explainer, Product Demo, Social Reel, Animation, Cinematic, Tutorial, Testimonial.
> Note: This generates a storyboard and script — not an actual video file. Video rendering via AI providers like Runway ML or Pika is on our roadmap.
6. AI Music Composition (Tracks)
What it does: Describe what you hear — "Upbeat electronic jingle for a tech product launch, 120 BPM, key of G" — and Studio generates a composition plan with structure, mood analysis, instrument suggestions, and a waveform visualization.
AI Model: GPT-4o-mini Token Cost: 30 tokens per generation Output: Composition metadata with structure (intro/verse/chorus/bridge/outro), mood tags, instrument list, and waveform visualization data.
Token deduction is atomic: We use a database operation with row-level locking. If two team members generate content simultaneously, the system prevents overdraft — one succeeds, the other gets a clear "insufficient tokens" error (HTTP 402).
Why fixed token costs instead of per-API-call billing? Predictability. Your team knows exactly how many documents, images, or dashboards they can create this month. No surprises when the bill arrives.
The generation flow
Every content type follows the same 5-step pipeline:
1. Create — User fills in a title, selects a type/style/genre, and optionally writes a prompt 2. Validate — Zod schema validates all input fields (type-safe at the API boundary) 3. Authorize — Check workspace membership, role, and feature entitlement 4. Deduct tokens — Atomic token consumption before any AI call 5. Generate — Call OpenAI (GPT-4o-mini or DALL-E 3), persist results, update entity status
If you provide a prompt during creation, generation starts automatically in the background — you see the detail page immediately, and results appear when ready.
If you skip the prompt, you can trigger generation later using the Generate with AI button on any detail page.
Reusable components
Studio ships with reusable components that make the generated content interactive:
Chart Renderer — Renders 8 chart types (bar, line, pie, area, scatter, radar, funnel, heatmap) with dark theme and custom tooltips
Chart Export — Export any rendered chart as PNG (2x retina) or SVG
Dashboard Renderer — Responsive grid layout for stat/chart/text/table widgets
Waveform Visualizer — Canvas-based audio waveform with play/pause animation
Video Storyboard — Timeline visualization for scene-by-scene storyboards
AI Generate Button — Reusable AI trigger with loading state, error handling, and token gating
Universal Actions — CRUD actions (edit, delete, confirm) for all entity types
Entitlement gating
Not every workspace has access to every module. Studio uses a layered entitlement system:
Plan-based: Free plans get limited access; paid plans unlock more modules
Owner grants: Platform admins can manually unlock modules for specific workspaces
Feature keys: `office`, `graph`, `tracks`, and `studio` (the parent key)
If a workspace doesn't have the required entitlement, they see a locked state with an upgrade prompt — no broken pages, no mysterious errors.
Internationalization
Every Studio page, component, and error message is fully translated in English and Indonesian (Bahasa Indonesia). This includes:
Sidebar navigation with collapsible sections
All 6 content type labels, descriptions, and form fields
Style/genre/format enum labels (10 image styles, 8 video styles, 12 music genres)
Generation status messages and error states
Token balance warnings and 402 error messages
What's on the roadmap
Coming soon
AI video rendering — Integration with Runway ML or Pika for actual video file generation
AI music synthesis — Integration with Suno API for actual audio file generation
Document export — PDF, DOCX, and XLSX export from structured JSON documents
Background generation queue — Async job processing for long-running AI tasks
Workflow execution engine — Actually run Tracks workflows with triggers and actions
Template marketplace — Share and discover templates across workspaces
Chart data connectors — Pull live data from databases, APIs, and CSV files
Try it today
Gigaviz Studio is available to all workspaces with an active plan. Start generating:
1. Navigate to Studio from your workspace sidebar 2. Pick a module — Office, Graph, or Tracks 3. Click Create New and write a prompt 4. Watch AI generate your content in seconds
Every workspace starts with a token balance. Once you see what Studio can do, you'll wonder how you ever created business content the old way.
---
*Gigaviz Studio is part of the Gigaviz platform — a multi-tenant SaaS ecosystem for business teams. Get started free or explore our pricing.*
