PhotoGPT AI Reviews

I just started using it 2-3 days ago, even when I got it like 2 months ago cause I didn't have the chance before.

I'll say I like it and a lot so far, I think the output options are great, quality, accuracy, and the better images I upload and the more posiive/negative feedback and prompts, it gives a much better bigh-quality output.

Will recommend it to anyone.

Software is a work in progress. Doesn't work well if multiple people need to be in a picture, they all end up looking like the subject. LLM has a hard time following directions.

Twist is that you get Additional aspect ratios only from tier 2. No use of getting Tier 1 if you want images for multiple platforms or aspect ratios.

Read my Full Review Here:

➡️ https://nimb.ws/Gw7zhjR

😛 Pros :

Transforms Selfies into Pro Headshots:No need for expensive studio shoots — just upload a few selfies.

100+ Themed Templates:Instantly generate images for various niches like fitness, fashion, holidays, or corporate branding.

AI Human Models Available:Create professional images even without using your own face — ideal for marketers, bloggers, and shy users.

Text-to-Image with Prompt Assistant:Turn creative ideas into visuals easily using built-in AI prompt generation.

Optimized for Social Media:Generates images with the correct aspect ratios for platforms like Instagram, LinkedIn, and more.

High-Resolution Output (in higher tiers):Perfect for printing or using in high-quality campaigns.

🙏 Cons?

Limited High-Resolution Upscaling in Lower Tiers: Only available in higher plans, which may require an upgrade for print-quality images.

Requires Multiple Selfies to Train AI Model: For best results, you need to upload 10–20 good selfies with various angles and expressions.

No Real-Time Photo Editing Tools: Unlike Photoshop or Canva, you can’t manually tweak or fine-tune images after generation.

✅ PROS
1) Highly realistic images – Many of the generated photos look incredibly lifelike, which is impressive for an AI-based tool.

2) User-friendly interface – The platform is clean, intuitive, and easy to navigate, which makes the experience smooth even for beginners.

❌ CONS
1) Lack of guidance for training photos – There's no clear instruction on what kind of images should be uploaded to train the model effectively (e.g. positioning, framing, body parts, etc.). This makes it difficult to know how to get the best results.

2) Hair generation issues – I don’t have hair, but the AI consistently generates images where I do. Even when explicitly stating this in the prompt, the result is often ignored.

3) Lack of resemblance – On many photos, I simply don’t recognize myself at all. The model seems to struggle with facial consistency.

4) Body shape inaccuracies – The AI appears to use generic templates rather than reflecting my actual body. There’s no consistency across images – my body looks completely different in each one.

💡 SUGGESTIONS FOR IMPROVEMENT
1) Provide clear photo submission guidelines – Add detailed instructions or examples for the initial training photos to ensure the AI can create the most accurate outputs possible.

2) Add custom correction parameters – Allow users to save persistent instructions for the model (e.g. “no hair”) so they don’t have to repeat the same details every time.

3) Improve support for bald users – The model should be trained to better handle users with no hair, respecting both the training data and textual prompts.

I uploaded 15 pictures of myself to train the model. That cost me 1,500 credits out of the 2,000 credits I have in Tier 1. But the results were really poor. So I wanted to upload a few more photos to improve the results. But to do that, I would have to retrain the model, which would cost me another 1,500 credits, which I no longer have. So I would have to wait a month just because I want to use a few additional images for my model. Unfortunately, that is not acceptable. I will probably return the tool.