Ideogram

  • Ideogram — Text-In-Image Specialist Image Generator (Ideogram Inc., Active, Not Shut Down)

    Not shut down.​ Ideogram (ideogram.ai) founded 2023 by ex-Google Brain researchers (CEO Mohammad Norouzi et al.), independent, active 2026-08, no acquire/rebrand/shutdown. Category ai-productivity-data; pipeline_stage none. Killer differentiator = readable embedded text in generated images​ — Ideogram 4.0 (Jun 2026, 9.3B open weights, commercial license) hits ~90–97% in-image OCR accuracy vs Midjourney ~30–40%, with bounding-box layout control + structured JSON prompting + Magic Fill inpaint + Magic Expand outpaint + Character Reference + Remix + 2K native + transparency; not Midjourney (art-first, text glitches), not Recraft (vector/SVG design-first), not DALL·E/GPT Image (bundled, weaker typography), not FLUX (open-weight photoreal, no hosted typography workflow). 2026 trap: subscription priority credits expire monthly (no rollover), 4.0 burns ~4× credits vs 2a Turbo, free tier = 10 slow credits/week + images permanently public/undeletable, Magic Fill stuck on 2.0/3.0 not 4.0, faces/photorealism trail Midjourney/FLUX, no Android native app, API billed separately ($0.025–$0.10/img) and 5× costlier than subscription at volume, very long strings still glitch.

    1. One-line positioning

    Ideogram = the “words actually spell right” image model: open ideogram.ai or iOS app → prompt “neon sign saying OPEN 24 HOURS” → pick Ideogram 3.0/4.0 (4.0 open-weight self-hostable) → Magic Prompt expands, bounding box pins text region, JSON for programmatic batches → output 2K PNG with transparency → Magic Fill edit region / Magic Expand uncrop / Remix variants / Character Reference lock a mascot → batch CSV on Pro → same weights via REST API + MCP server for Cursor/Claude Code. Fits ai-productivity-data; pipeline_stage none (prompt→layout→render→edit→batch→ship). For poster/ad/social/logo/packaging/print-on-demand teams where text must be legible, it’s the only production-grade seat; for painterly art, editorial photoreal, or SVG brand systems, Midjourney/FLUX/Recraft win.

    2. Core features

    Layer Capability Notes
    Models Ideogram 2.0 / 3.0 / 4.0 (Jun 2026, 9.3B, open weights, commercial license) 4.0 = ~4× credit burn vs 2a Turbo; weights on HF/GitHub
    Text rendering In-image OCR ~90–97% (4.0), Latin + major EU langs, font/style control Still glitches on 10+ word paragraphs
    Layout control Bounding-box element placement, structured JSON prompting Enables reproducible programmatic gen
    Editing Magic Fill (inpaint, 2.0/3.0 only), Magic Expand (outpaint), Upscale to 8K, BG remove, transparent PNG 4.0 not in Magic Fill yet
    Consistency Character Reference, Style Reference / Style Codes Pro/Team/Ent
    Iterate Remix, Describe (image→prompt), Magic Prompt auto-expand, negative prompts
    Batch CSV batch gen (Pro+), concurrency 8 (Plus) / 32 (Pro)
    Output Native 2K, JPG (free) / PNG (paid), aspect + custom dims
    Distribute Web app + iOS app (4.8★) + REST API + MCP server + ComfyUI day-0 No Android native app
    Enterprise Custom model training on brand assets, private credits, API volume discount, priority support Min ~1M img/mo

    3. Pricing (2026-08, USD — web credits ≠ API)

    Tier Monthly Annual (eff/mo) Priority credits Key gates
    Free $0 $0 10 slow/wk (~40 imgs) Public+undeletable, JPG, 1 concurrent, no private/Magic Fill/upscale
    Plus $20 $15 1,000/mo Private gen, PNG, 8 conc, unlimited slow, char ref, Magic Fill/Expand, top-up $4/150cr
    Pro $60 $42 3,500/mo 32 conc, batch CSV, largest queue, top-up $4/250cr
    Team $30/seat $20/seat 1,500/seat Central billing, collab, min 2 seats
    Enterprise Custom Custom Private custom models, API volume, priority support
    API PAYG 2a Turbo $0.025/img · 3.0 Turbo $0.04 · Default $0.08 · Quality $0.09 · 4.0 ~$0.10 · edits $0.10–0.12

    pricing_type = freemium-credit-subscription-plus-payg-api; starting_price = $0; free_tier = "yes; 10 slow credits/week (~40 imgs), public+undeletable gallery, JPG only, 1 concurrent, Magic Prompt/Remix/styles only, no private gen/Magic Fill/upscale/PNG/char-ref"; pricing_note = "ideogram-inc-independent-active-2026-no-shutdown; sub-priority-credits-expire-monthly-no-rollover-topups-roll-over; 4.0-4x-credit-burn-vs-2a-turbo; plus-20-or-15yr-pro-60-or-42yr-team-30-or-20yr-seat; api-separate-0.025-0.10-img-5x-costlier-than-sub-at-volume; free-images-permanent-public-undeletable; magic-fill-not-on-4.0"

    4. Access Type

    access_type: web-app-plus-ios-app-plus-rest-api-plus-mcp

    access_display: 🌐 ideogram.ai · 📱 iOS app (4.8★, no Android native) · 🔌 REST API (Generate/Remix/Edit/Reframe/ReplaceBG/Inpaint) + MCP server (Cursor/Claude Code) · ComfyUI node · no desktop, no Discord bot, no Figma plugin first-party

    5. Reviews (2026)

    Source Read
    AIVario / AIUnpacking “Text accuracy 90–95% vs MJ 30–40%; Plus $15/yr the designer sweet spot; 4.0 credit burn surprises heavy users”
    SurePrompts “Specialist, not generalist — pair with Midjourney/FLUX; faces/photoreal trail MJ”
    Delv/UsagePricing “Credit expiry + 4.0 4× burn + API 5× sub cost = budget carefully; open weights save self-hosters”
    ToolChase 4.2/5 — “poster/social/logo workflow unbeatable on text; free tier useless for client work (public+undeletable)”

    Praise: only production-grade text-in-image​ (signs/posters/T-shirts/packaging), Magic Prompt lowers prompt skill floor, Character Reference locks mascots across a campaign, 4.0 open weights = auditable + self-hostable, Plus $15/yr cheaper than Midjourney private tier, API+MCP+Cmdfy fits dev pipelines, iOS app solid.

    Gripes: priority credits evaporate monthly​ (top-ups only rescue), 4.0 burns 4× credits silently, free images permanent public + can’t delete, Magic Fill not on 4.0 (edit quality regresses if you want 4.0 base), faces/photoreal still behind Midjourney v7/FLUX 2, very long strings typo, no Android app, API 5× pricier than subscription at same volume, slow queue 20 min peak on free.

    6. Best for / Not for

    Best for Not for
    Marketers shipping ad-creative variants with embedded headlines (batch CSV) Editorial/art-direction photographers (Midjourney/FLUX)
    Designers doing posters/flyers/event signage/book covers with real copy Vector/SVG brand-system work (Recraft)
    Print-on-demand sellers (typographic tees/posters/greeting cards) Teams needing strict commercially-licensed training data (Adobe Firefly)
    Packaging/ecom mockups showing readable copy pre-print Casual ChatGPT users happy with GPT Image bundled
    Devs embedding text-accurate gen in apps (API+MCP, or self-host 4.0) High-volume production that forgets credit expiry → API cost shock

    7. Competitors

    Tool Lane Starts
    Midjourney v7 Artistic/photoreal king, text weak $10/mo
    FLUX 2 (BFL) Open-weight photoreal + decent text API $0.02–0.08
    Recraft v3 Vector/SVG + brand kits + text $12/mo
    DALL·E / GPT Image Bundled in ChatGPT, conversational $20/mo
    Adobe Firefly Commercial-safe, CC integration Bundled/$10-mo
    Leonardo.Ai Game/assets/custom models $10/mo
    Stable Diffusion 3.5 Self-host, max control, text variable $0
    Krea / Nano Banana Pro Real-time / photoreal+text $10-mo / free

     

Do Not Sell or Share My Personal Information Cookie Settings