OpenAI's previous generation image model with higher quality than DALL-E 2 and support for larger resolutions
Settled by community votes across 1 shared challenge, with an AI judge weighing in on each.
DALL-E 3
#35 of 44 in Text-to-Image
Not enough comparable category data
The chart appears once both models have ratings across at least three shared arena categories.
HiDream I1 Fast
#38 of 44 in Text-to-Image
Where the votes landed
DALL-E 3
0.0%
win rate
Ties
0.0%
HiDream I1 Fast
100.0%
win rate
Challenge by challenge
The strongest take from each model on every shared challenge, with the AI judge's read.
Fantasy Warrior
Text-to-Image“Close portrait of a battle-worn paladin in ornate engraved plate armor, hair braided with small beads, faint scars and dirt on the skin, warm torchlight reflecting off metal, shallow depth of field, bokeh sparks, lifelike eyes, highly detailed texture on leather straps and cloth underlayer.”
AI Judge Analysis
DALL-E 3
- + Excellent execution of warm torchlight reflecting off intricate engravings.
- + Very lifelike eyes with realistic wetness and depth.
- + Dynamic composition with a cinematic bokeh spark effect.
- − The braids are somewhat obscured by the helmet compared to the other model.
HiDream I1 Fast
- + Clearly visible and well-executed braids with color-coordinated beads.
- + Good focus on the leather straps and the transition between different armor materials.
- + Effective use of warm background lighting to create mood.
- − The scars look like surface paint/bruising rather than integrated skin texture.
- − The facial lighting is a bit flat compared to the metal textures.
Verdict: DALL-E 3 succeeds by creating a more cohesive and artistic portrait where the lighting truly feels like it is reflecting off the various surface textures of the plate armor. HiDream I1 Fast does a better job of showcasing the requested braids and beads, but it falls short on the realistic rendering of the scars and the lifelike quality of the skin compared to DALL-E 3.
Explore each model
Distilled version of HiDream AI's 17B parameter text-to-image model