BUYER PROMPT / PUBLISHED EVIDENCE
Create a reproducible benchmark for AI image generators covering people, products, scenes, text, composition, references, edits, and difficult prompts.
An exact prompt-level view of which reviewed brands surfaced and which public URLs the configured AI models cited. No answer text or tenant data is published.
- Intent
- commercial
- Model providers
- 3
- Categories
- 1
- Last observed
- Jul 27, 2026, 12:00 AM UTC
COMPETITIVE OUTCOME
Brands surfaced
Ordered by provider breadth, then total mentions and owned-domain citations. A mention is not an endorsement or a position claim.
| # | Brand | Category | Providers | Mentions | Owned citations | Provider evidence |
|---|---|---|---|---|---|---|
| 01 | Midjourneymidjourney.com ↗ | Image AI | 2 | 3 | 0 | Anthropic · Gemini |
MODEL COMPARISON
Answer matrix
One row per configured provider and category snapshot. Hashes prove answer identity without publishing stored answer text.
| Provider | Model version | Category | Brands surfaced | Citations | Unique domains | Answer hash |
|---|---|---|---|---|---|---|
| Openai | openai/gpt-4o-mini | Image AI | None observed | 4 | 4 | cdc7bcd8…0a154 |
| Anthropic | anthropic/claude-haiku-4-5 | Image AI | Midjourney | 5 | 5 | c4180946…e31df |
| Gemini | google/gemini-2.5-flash | Image AI | Midjourney | 8 | 8 | 5da03de6…92e2e |
URL-DERIVED CITATIONS
Sources cited
Hostnames are derived from the returned public URLs. Best position is the smallest citation position observed.