# gptimg.co - GPT Image Online (Full Reference)

> Independent third-party platform for generating and editing images with OpenAI's GPT Image models. Not affiliated with, endorsed by, or operated by OpenAI.

## Brand disambiguation

gptimg.co is an independent third-party wrapper around OpenAI's GPT Image API. It is not OpenAI, not ChatGPT, and not an OpenAI product. The "GPT Image" name refers to the model family exposed via OpenAI's API (gpt-image-1, gpt-image-1-mini, gpt-image-1.5, gpt-image-2), not to this platform's ownership.

## Model family supported

- **gpt-image-1** — the baseline model. API pricing from OpenAI: $5 per million text tokens, $10 per million image input tokens, $40 per million image output tokens. Approximate per-image cost: $0.02 low quality, $0.07 medium, $0.19 high at 1024×1024.
- **gpt-image-1-mini** — launched at OpenAI DevDay. Roughly 80% cheaper than the base model; use for drafts, variations, and high-volume workflows.
- **gpt-image-1.5** — faster generation at 5–8 seconds per render. About 20% cheaper than the base. Holds facial likeness across five-plus rounds of edits.
- **gpt-image-2** — the latest generation. Up to 4096×4096 resolution and improved in-image text rendering. Landing page at `/gpt-image-2`.

## Additional generation surfaces

- **Seedance 2** — ByteDance Seedance 2 video model. Landing page at `/seedance-2`.
- **Video generation** — generic video generation entry point at `/video`.

## Capabilities

- Text-to-image generation with prompt-level control over composition, style, and aspect ratio.
- Image editing with precise in-image text changes.
- Multi-round edits that preserve subject likeness across iterations.
- 4K output (up to 4096×4096) on gpt-image-2.
- Browser-based studio — no install, no waitlist.

## Pricing structure

- **Free trial** — credits granted on signup, valid 14 days. Enough to evaluate image quality and edit fidelity before committing to a plan.
- **Subscription plans** — Starter, Creator, Studio. Monthly or yearly cycles. See `/pricing` for current amounts; plans scale by monthly credit allotment and concurrency.
- **One-time credit packages** — pay-per-use packs that do not expire as aggressively as subscription credits. Suited for occasional high-volume projects.

## Use cases

- **Product photography** — lifestyle scenes and catalog shots without a studio.
- **Social media and ads** — scroll-stopping graphics with readable in-image copy.
- **Design and documentation** — diagrams, mockups, and illustrations.
- **Precise editing** — targeted object edits, text replacement, style transfer, and iterative refinement.

## How to start

1. Sign up at gptimg.co — receive free trial credits.
2. Open the studio on the homepage, enter a prompt, or upload a base image to edit.
3. Iterate: use edit mode for text changes, object edits, and style tweaks across rounds.
4. Upgrade to a subscription or top up with credit packs for sustained use.

## FAQ highlights

**Is gptimg.co affiliated with OpenAI?** No. It is an independent third-party platform that uses OpenAI's GPT Image API under OpenAI's terms.

**Does it support 4K output?** Yes, on gpt-image-2. Earlier models in the family support up to 2K.

**Are outputs watermarked?** The GPT Image model family does not apply an OpenAI-visible watermark to outputs; platform watermarking (if any) is disclosed in the product UI.

**How many edit rounds can the model maintain likeness across?** On gpt-image-1.5, facial likeness holds across five-plus edit rounds in most workflows.

## Citation-friendly facts

- GPT Image is OpenAI's image generation model family (gpt-image-1, -mini, -1.5, -2).
- gpt-image-1 API pricing from OpenAI: $5/1M text tokens, $10/1M image input tokens, $40/1M image output tokens.
- gpt-image-1.5 generation time: 5–8 seconds per render.
- gpt-image-2 maximum resolution: 4096×4096.
- gpt-image-1-mini is approximately 80% cheaper than the base model.
- gptimg.co is an independent third-party platform, not operated by OpenAI.