About image 2.5

What image 2.5 provides

image 2.5 is a browser workspace for AI image generation, AI video generation and reading a prompt out of an image. You pick a model, describe what you want, and the result returns to the same page.

• Image generation models listed today: GPT Image 2.5, GPT Image 2, Nano Banana Pro, Nano Banana 2 and Nano Banana 2 Lite.
• Video generation families listed today: Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast and Seedance 2.0 Mini.
• Image to prompt, which reads a structured prompt out of a picture you upload.
• A personal centre holding your account information, membership and credit balance, a credit ledger, your past creations and your order records.

Choose the right image model

Each published model page combines an explanation, sourced examples, prompts and a workspace already set to that model.

Use GPT Image 2.5 and GPT Image 2 for OpenAI image generation, or choose Nano Banana Pro, Nano Banana 2 and Nano Banana 2 Lite from the Google model group.

The general image generator keeps every available model in one selector when you want to compare options without leaving the workspace.

Create images and videos in one workspace

The image and video tools keep the prompt, model controls, reference uploads, quote and result on the same page.

• Image generation supports text prompts, image editing and reference-led creation when the selected model exposes those modes.
• Video generation provides separate Seedance pages for text, image and supported reference workflows.
• The workspace shows the current credit quote before submission and returns the task result without sending you to another tool.

What the workspace keeps for you

The workspace is built to survive a reload. Task state is restored rather than lost, and the site remembers the model you used last.

• Earlier tool runs are listed in your browser and tied to your account, so a repeat request does not have to be rebuilt from memory.
• Membership, credit balance, credit ledger and order records are read from the account and payment services when you open the personal centre.
• Ten languages are maintained in full: English, Simplified Chinese, Traditional Chinese, Japanese, Spanish, German, Persian, Arabic, Russian and French. Each carries its own copy rather than falling back to English.

What the site does not do

The current scope is image generation, video generation and image to prompt. Chat, music and audio tools are not part of it, and the site does not advertise features that are not wired up.

• No promise that a generation will look like the example on a page. Examples are credited to their source and are not a commitment about your result.
• No removal of watermarks or ownership marks, and no use of a customer upload as a public sample.
• Prices are read from the payment service catalogue rather than typed into these pages, and the amount charged is resolved when checkout opens.

Getting help

Support runs through one address, support@image25.top, and you can write in English, Simplified Chinese, Traditional Chinese or Japanese.

A message that names the account email, what you were doing, the model and settings, and any task or order identifier can usually be answered without a second round of questions. If a generation failed, the task identifier is the fastest way to find the record.

Before writing, the guide pages, the individual model pages and the pricing page answer most questions about credits and model behaviour.