模型广场

New unified architecture combining generation and editing; knowledge-based output and reference consistency; up to 4K.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Broad upgrade via model scaling; sharper subject detection, stricter reference fidelity, stronger dense text rendering.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Adds deep reasoning; better interprets complex prompts with common-sense knowledge; stronger edit consistency.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Built for fast generation and conversational editing; low latency and cost; multimodal text-and-image input.

上下文

33K

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Mid-tier image generation and editing model in the series, with near-flagship visual quality at higher generation speed. Renders legible text and photorealistic subjects, suited for infographics and marketing visuals.

上下文

131K

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

The lightest image model in its series, producing images in about four seconds with consistent character rendering and precise edits. Suited for high-throughput pipelines and interactive visual iteration.

上下文

66K

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Built-in reasoning; complex multi-turn creation; up to 4K; optional web grounding.

上下文

66K

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Natively multimodal; accepts text and image input, outputs images; supports generation and editing.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Lower-cost variant of the same architecture; same text-and-image I/O; lower per-call cost.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Stronger instruction following and edit precision; better preserves faces and logos; faster generation.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Flexible sizing and high-fidelity reference inputs; faster generation and editing; stronger multilingual text and complex layouts.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Offering state-of-the-art visual quality, precise instruction following, and support for large-scale batch processing.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Unifies generation and editing in one model; native 2K; professional typography and infographics.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Fuses generation and editing; more professional text rendering and richer realistic textures and scenes.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Stronger industrial design, geometric reasoning, and character consistency; text, object, and style edits.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Multi-image input/output and custom resolution; multiple edited results per request.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-

Diverse styles; multi-line layouts and paragraph-level text generation.

上下文

待核

输入 / 1M tokens

-

输出 / 1M tokens

-

缓存读 / 1M

-

缓存写 / 1M

-