T2I Evaluation

3 Prompts

6 models

Rating

Save this link to find your results later

Download images

Images remain available for at least 5 years after creation.

Generating → Launching eval

Crowdsourced evaluation in progress

1 / 6 sessions completed

Preliminary results (1/6 sessions) — these will update as more ratings come in.

Elo scores

Each bar spans a 95% credible interval; the dot marks the mean. Hover for exact values.

Cost vs. Elo score

Generated images

eggs with legs

beans in jeans

frog on a log that looks like a hog in a smoggy bog

“eggs with legs”

${'text': 'eggs with legs', 'image_s3_key': None} — MiniMax Image$

MiniMax Image

${'text': 'eggs with legs', 'image_s3_key': None} — Flux.1 Schnell$

Flux.1 Schnell

${'text': 'eggs with legs', 'image_s3_key': None} — GPT Image 1 Mini$

GPT Image 1 Mini

${'text': 'eggs with legs', 'image_s3_key': None} — Riverflow V2 Fast$

Riverflow V2 Fast

${'text': 'eggs with legs', 'image_s3_key': None} — Kling V3$

Kling V3

“beans in jeans”

${'text': 'beans in jeans', 'image_s3_key': None} — MiniMax Image$

MiniMax Image

${'text': 'beans in jeans', 'image_s3_key': None} — Flux.1 Schnell$

Flux.1 Schnell

${'text': 'beans in jeans', 'image_s3_key': None} — GPT Image 1 Mini$

GPT Image 1 Mini

${'text': 'beans in jeans', 'image_s3_key': None} — Riverflow V2 Fast$

Riverflow V2 Fast

${'text': 'beans in jeans', 'image_s3_key': None} — Kling V3$

Kling V3

“frog on a log that looks like a hog in a smoggy bog”

${'text': 'frog on a log that looks like a hog in a smoggy bog', 'image_s3_key': None} — MiniMax Image$

MiniMax Image

${'text': 'frog on a log that looks like a hog in a smoggy bog', 'image_s3_key': None} — Flux.1 Schnell$

Flux.1 Schnell

${'text': 'frog on a log that looks like a hog in a smoggy bog', 'image_s3_key': None} — GPT Image 1 Mini$

GPT Image 1 Mini

${'text': 'frog on a log that looks like a hog in a smoggy bog', 'image_s3_key': None} — Riverflow V2 Fast$

Riverflow V2 Fast

${'text': 'frog on a log that looks like a hog in a smoggy bog', 'image_s3_key': None} — Kling V3$

Kling V3

Navigation

Legal

T2I Evaluation

Elo scores

Cost vs. Elo score

Generated images