/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

← → days · ↑ ↓ browse · Enter similar · o open

Nano Banana Pro is great at following instructions, generates interim “thought images”, and makes full infographics with well-rendered text from a short prompt

Simon Willison / Simon Willison's Weblog :

Simon Willison's Weblog Simon Willison

Context & Ripple Effects

This report captures the early capability case for Nano Banana Pro: short prompts can produce instruction-following visual work, including text-heavy infographic layouts. The next day’s coverage framed the product as Gemini 3 Pro Image with greater control and improved text rendering in the Gemini app.

Later releases extended the same arc: Nano Banana 2 emphasized faster generation and precision text rendering, while Nano Banana 2 Lite pushed image output toward lower-cost, four-second delivery. That makes the initial focus on prompt control and legible text consequential beyond image aesthetics.

First-order effects

  • Users can turn brief instructions into more complete visual deliverables, reducing the manual iteration otherwise needed to assemble an infographic and repair generated text.
  • Nano Banana Pro’s reported interim “thought images” make the generation process more inspectable, while its Gemini-app availability puts those capabilities in front of general users immediately.

Second-order effects

  • Design and content workflows can shift from selecting isolated generated images to directing and reviewing a model’s composed visual output; reliable text rendering is the key enabler for that shift.
  • Competing image-model providers face pressure to improve controllability and in-image typography, not just visual quality—capabilities Google explicitly foregrounded in its Gemini 3 Pro Image launch.

Third-order effects

  • If instruction fidelity and typography continue to improve alongside the later speed and cost reductions, image generation will increasingly compete in production-oriented communication tasks rather than only illustration.
  • The durable differentiation may move toward the editorial control layer: tools that preserve brief, review, revision, and approval context around generated visual assets.

The trend: Generative-image products are evolving from prompt-to-picture tools into controllable visual-production systems that can produce usable, text-bearing communication assets.

Discussion

  • @simonw Simon Willison on x
    Nano Banana Pro, released this morning, is clearly the best image generation model. Superb instruction following, plus it can generate full infographics (with correct spelling and properly rendered text!) from a short prompt based on running extra searches https://simonwillison.n…