GPT Image 2 各版本与参数
| 参数 | GPT Image 2 |
|---|---|
| 时长 | — |
| 画面比例 | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 3:2 · 2:3 · 4:5 · 5:4 · 21:9 · 9:21 · 1:2 · 2:1 · 1:3 · 3:1 |
| 分辨率 | — |
| 原生音频 | — |
| 图生视频 | — |
| 参考图生视频 | — |
| 积分/秒 | — |
| 5 秒片段消耗 | 2 cr/image |
什么是 GPT Image 2?
GPT Image 2 is OpenAI's production image generation model and the strongest option on FlyAIgh for any image where readable text is part of the deliverable — posters, ad creatives, product labels, slide graphics, infographics. Where most image models render text as noisy lookalike glyphs, GPT Image 2 produces clean letterforms that match what you typed in the prompt.
Pricing is unusually aggressive at 2 credits per image — the second-cheapest image option on FlyAIgh after Z Image Turbo (1 credit). The Free tier's 15 credits cover seven full GPT Image 2 generations.
Both text-to-image and image-to-image (edit) modes are supported. In edit mode you supply up to 4 reference images plus a text instruction and the model produces a modified version. Output supports five aspect ratios (1:1, 16:9, 9:16, 4:3, 3:4) in HD quality.
One small caveat: the i2i (edit) endpoint doesn't accept the quality parameter that text-to-image does — the picker is reduced to a single 'Auto' option in edit mode so you don't select something that gets silently dropped.
GPT Image 2 is particularly strong at faces, hands, on-image text, structured layouts and any prompt with explicit lettering ('a poster that reads ...'). It is weaker than Midjourney v7 for distinctive artistic style and weaker than Nano Banana Pro for 4K detail. Pick GPT Image 2 when text legibility, editing capability and cost-per-image matter most.
On FlyAIgh, GPT Image 2 ships under the same credit wallet as every other model. No separate OpenAI subscription or API key.
GPT Image 2 vs Nano Banana vs Midjourney v7
| 能力 | GPT Image 2 | Nano Banana | Midjourney v7 |
|---|---|---|---|
| 原生音频 | |||
| 图生视频 | |||
| 参考图生视频 | |||
| 首末帧 |