cff-version: 1.2.0
title: "AI Design Benchmark"
message: "If you use this data, please cite it and link the FontVibe article of the benchmark you used, e.g. https://fontvibe.ai/blog/gpt-6-1-terra-vs-claude-gemini-design-test."
type: dataset
authors:
  - name: "FontVibe"
    website: "https://fontvibe.ai"
abstract: >-
  Open data from FontVibe's benchmarks of AI models doing design work with
  text. 2026-10-llm-designer: eight language models (GPT-6.1 Terra, GPT-5.6
  Terra, GPT-6.1 Sol, GPT-6 Luna, Claude Opus 5.5, Claude Sonnet 5.5, Gemini 4
  Argon, Gemini 3.8 Flash) act as the designer in a fixed design pipeline, on 12
  briefs in four layers and six languages, three runs each (288 runs, 269 ranked
  designs plus 18 side-note reruns), scored blind. Two image-model benchmarks (Grok Imagine 2.0, Qwen Image 3.0) compare
  text rendering on ten design jobs. Every brief, output, image link, score and
  written reason is included; a script recomputes the published tables of the
  llm-designer benchmark.
keywords:
  - benchmark
  - large language models
  - graphic design
  - typography
  - text-to-image
  - evaluation
  - AI design
license: CC-BY-4.0
url: "https://fontvibe.ai/blog/gpt-6-1-terra-vs-claude-gemini-design-test"
repository-code: "https://github.com/Funovate-Technology/ai-design-benchmark"
doi: "10.5281/zenodo.23803802"
version: "1.0.1"
date-released: "2026-10-31"