InstaFame 0.5 vs 1.0: A 20-Image Blind Test
InstaFame strengths 0.5 and 1.0 tested across 20 matched Krea 2 scenes with the same Mila identity LoRA, prompts, seeds, sizes, and settings.
Read guide →Model research hub
Controlled AI image model comparisons
Choose by face preservation, scene accuracy, and completion. Our July 2026 five-prompt portrait test records 25 outputs with one frozen input; it is not a universal quality ranking.
Curated guides
Choose a model test
Comparing FLUX.1 [dev] vs FLUX.2 [dev] for Character LoRAs
FLUX.1 and FLUX.2 character LoRAs tested with the same 46-photo Mila dataset, three exact prompts, matched seeds, and 3:4 framing.
Read guide →
Seedream V5 Pro Wins, but 4.5 Is Close
V5 Pro narrowly beat Seedream 4.5 across 19 valid tests. V5 Lite trailed.
Read guide →
One Prompt, Five Models: Blind Image Test
Vote through five same-input comparisons across Seedream 4.5, GPT Image 2 Medium, Grok, Seedream V5 Lite, and Seedream V5 Pro, then reveal live results.
Read guide →
Best AI Image Model for Realistic Mirror Selfies: Blind Test
Controlled 18-prompt benchmark: Nano Banana 2 generated 83% of attempts versus 67% for Nano Banana and Nano Banana 2 Lite, with 14 blind visual comparisons.
Read guide →
One Prompt, Five Models: Controlled Same-Input Tests
Five exact prompts tested on one Mila input across Seedream 4.5, GPT Image 2 Medium, Grok, Seedream V5 Lite, and Seedream V5 Pro.
Read guide →
Which AI Image Generator Blocks More? A Controlled 3-Prompt Test
Nano Banana 2, GPT Image 2 Medium, and Seedream 4.5 tested with one Mila input: 9 attempts, 7 images, and 2 recorded GPT safety blocks.
Read guide →Seedream 4.5 vs V5 Lite vs V5 Pro: Same-Input Test
Five exact prompts tested with one Mila input across Seedream 4.5, V5 Lite, and V5 Pro.
Read guide →
Seedream V5 Lite vs Pro: Same-Input Portrait Test
Five exact prompts tested with one Mila input across Seedream V5 Lite and V5 Pro.
Read guide →Test coverage and methods
Each current comparison uses one exact input for every model attempt in a row. Prompt hash, input key and hash, profile, aspect ratio, run ID, exact model ID, request fingerprint, and output hash are verified before publication.
At a glance
Benchmark coverage
| Test | Models covered | What is controlled |
|---|---|---|
| InstaFame strength blind test | InstaFame 0.5 and InstaFame 1.0 on Krea 2 Turbo | 20 scenes; one frozen Mila identity LoRA at 1.0; matched prompts, seeds, sizes, endpoint, and settings; one blind reviewer |
| FLUX LoRA migration pilot | FLUX.1 Dev LoRA and FLUX.2 Dev LoRA | Exact 46-photo training archive; 1,000 training steps; 3 matched prompts and seeds; 6 outputs; identity review |
| Explicit Seedream benchmark | Seedream 4.5, Seedream V5 Lite, Seedream V5 Pro | 20 prompts; frozen Mila or Lily input per row; 59 control-valid outputs; one preserved invalid control; one blinded reviewer |
| Realistic mirror-selfie blind test | Nano Banana 2, GPT Image 2 Medium, Grok, Seedream V5 Lite, Seedream V5 Pro | 1 exact prompt; one Lily input; 4 verified outputs; 1 visible block; anonymous reader vote |
| One prompt across five models | Seedream 4.5, GPT Image 2 Medium, Grok, Seedream V5 Lite, Seedream V5 Pro | 5 prompts; one Mila input; 25 verified outputs; blind reader vote; exact prompt, request, model, and output hashes |
| Adult-fashion moderation boundary test | Nano Banana 2, GPT Image 2 Medium, Seedream 4.5 | 3 prompts; 9 attempts; 7 images; 2 GPT safety blocks; no broader moderation ranking |
| Seedream version test | Seedream 4.5, V5 Lite, V5 Pro | 5 exact prompts and one input across all three versions |
| Seedream V5 Lite vs Pro | V5 Lite and V5 Pro | 5 priority prompts, including the kitchen portrait, with identical inputs per pair |
Explore related topic hubs
Ready to try a prompt with your own character?
Browse public prompt packs, choose a look, and remix it with an approved photo or trained character.
Explore prompt packs