GPT Image 1 Mini

GPT-image-1-miniは、テキストプロンプトから直接新しい画像を作成するための、OpenAIの軽量モデルです。最大1536×1024の解像度で、高速かつ手頃な価格の画像生成を提供し、品質と忠実度を調整できます。大量のクリエイティブ用途に最適ですが、微細なディテールの表現力と写実性はプレミアムモデルに比べて劣ります

モダリティ

API 利用例

curl --request POST "https://gptproto.com/v1/images/generations" \
  --header "Authorization: Bearer $GPTPROTO_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "model": "gpt-image-1-mini",
    "prompt": "a cat",
    "size": "1024x1024"
  }'

GPT Image 1 Mini の料金

1回のサンプル費用から始め、テスト予算を選びます。GPTProto 料金は表示価格より 30% お得です。

利用量

モデル

$0.0061 / 枚

月あたり画像数

シナリオプリセット

関連モデル

すべてのモデル

GPT Image 1 Mini

現在

byOpenAI$1.75/M input$5.6/M output

GPT Image 2.5 Sunburst

byOpenAI$7.6/M input$28.5/M output

GPT Image 2.5 Flare

byOpenAI$7.6/M input$28.5/M output

Grok Imagine Image 2.0

byGrok$0.04 / 1回あたり1K · 2K · more

Seedream 5.0 Pro (Build 260628)

byBytedance$0.0405 / 1回あたり1K · 2K

Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image)

byGoogle$0.0202 / 1回あたり1K

Nano Banana 2 (Gemini 3.1 Flash Image)

byGoogle$0.0402 / 1回あたり1K · 2K · 4K

GPT Image 2

byOpenAI$6.4/M input$24/M output

Nano Banana 2 (Gemini 3.1 Flash Image)

byGoogle$0.0402 / 1回あたり1K · 2K · 4K

Seedream 5.0 (Build 260128)

byBytedance$0.0298 / 1回あたり

Doubao Seedream 5.0 (Build 260128)

byBytedance$0.0298 / 1回あたり

Grok Imagine Image

byGrok$0.012 / 1回あたり

Kling Image o1

byKling$0.0224 / 1回あたり

GPT Image 1.5

byOpenAI$5.6/M input$22.4/M output

Grok Imagine 0.9

byGrok$0.135 / 1回あたり

Image Watermark Remover

byGPTProto$0.01 / 1回あたり

Image Zoom

byGPTProto$0.02 / 1回あたり

Qwen Image

byQwen$0.0315 / 1回あたり

Flux Kontext Max

byFlux$0.064 / 1回あたり

Flux Kontext Pro

byFlux$0.032 / 1回あたり

Ideogram Reframe v3

byIdeogram$0.048 / 1回あたり

Ideogram Edit v3

byIdeogram$0.048 / 1回あたり

Ideogram Remix v3

byIdeogram$0.048 / 1回あたり

GPT Image 1

byOpenAI$7/M input$28/M output

Midjourney

byMidjourney$0.0608 / 1回あたり

モデル			解像度	入力 → 出力
GPT Image 1 Mini現在 $		$1.75 / $5.60 per 1M	—	入力: テキスト入力: 画像出力: 画像
GPT Image 2.5 Sunburst $		$7.60 / $28.50 per 1M	—	入力: テキスト入力: 画像出力: 画像
GPT Image 2.5 Flare $		$7.60 / $28.50 per 1M	—	入力: テキスト入力: 画像出力: 画像
Grok Imagine Image 2.0 $	—	$0.04 / 1回あたり	1K · 2K · more	入力: テキスト入力: 画像出力: 画像
Seedream 5.0 Pro (Build 260628) $		$0.04 / 1回あたり	1K · 2K	入力: テキスト入力: 画像出力: 画像
Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image) $		$0.02 / 1回あたり	1K	入力: テキスト入力: 画像出力: 画像
Nano Banana 2 (Gemini 3.1 Flash Image) $		$0.04 / 1回あたり	1K · 2K · 4K	入力: テキスト入力: 画像出力: 画像
GPT Image 2 $		$6.40 / $24.00 per 1M	—	入力: テキスト入力: 画像出力: 画像
Nano Banana 2 (Gemini 3.1 Flash Image) $		$0.04 / 1回あたり	1K · 2K · 4K	入力: テキスト入力: 画像出力: 画像
Seedream 5.0 (Build 260128) $		$0.03 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
Doubao Seedream 5.0 (Build 260128) $		$0.03 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
Grok Imagine Image $		$0.01 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
Kling Image o1 $		$0.02 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
GPT Image 1.5 $		$5.60 / $22.40 per 1M	—	入力: テキスト入力: 画像出力: 画像
Grok Imagine 0.9 $	—	$0.14 / 1回あたり	—	入力: テキスト出力: 画像
Image Watermark Remover $	—	$0.01 / 1回あたり	—	入力: 画像出力: 画像
Image Zoom $	—	$0.02 / 1回あたり	—	入力: 画像出力: 画像
Qwen Image $		$0.03 / 1回あたり	—	入力: テキスト出力: 画像
Flux Kontext Max $		$0.06 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
Flux Kontext Pro $		$0.03 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像
Ideogram Reframe v3 $		$0.05 / 1回あたり	—	入力: 画像出力: 画像
Ideogram Edit v3 $		$0.05 / 1回あたり	—	入力: 画像出力: 画像
Ideogram Remix v3 $		$0.05 / 1回あたり	—	入力: テキスト出力: 画像
GPT Image 1 $		$7.00 / $28.00 per 1M	—	入力: テキスト入力: 画像出力: 画像
Midjourney $		$0.06 / 1回あたり	—	入力: テキスト入力: 画像出力: 画像

GPT Image 1 Mini API — 30% Off for Budget Image Generation

Use OpenAI's cost-efficient image model through GPTProto at 30% below OpenAI's token rates. Generate or edit square, portrait, and landscape images with text or image input, three quality tiers, one API key, and one balance shared across 200+ models. OpenAI has scheduled this model for API retirement on December 1, 2026, so it is best suited to current low-cost workloads with a planned path to GPT Image 2.

30% Below OpenAI Token Rates

GPTProto lists image input at $1.75/M tokens and image output at $5.60/M, 30% below OpenAI's $2.50/M and $8.00/M rates.

Generate and Edit Images

Send text prompts or image references and receive image output. Use the model to generate new visuals or revise supplied images within the GPT Image workflow.

Three Sizes and Quality Tiers

Choose 1024x1024, 1024x1536, or 1536x1024 output and set low, medium, high, or auto quality to control detail and cost.

A Clear GPT Image 2 Migration Path

OpenAI has deprecated this model and will remove it from its API on December 1, 2026. Use GPT Image 2 for new long-lived integrations.

What Is the GPT Image 1 Mini API?

GPT Image 1 Mini is OpenAI's lower-cost version of GPT Image 1. It is a natively multimodal image model: text is accepted as input, images can be supplied for editing, and the result is an image rather than a text response. The model string is gpt-image-1-mini.

The model is designed for workloads where generation cost matters more than maximum detail. Typical GPT Image 1 Mini API use cases include prompt testing, storyboard drafts, product concepts, educational illustrations, social-media drafts, and batch assets that will be refined later. It supports low, medium, and high quality at 1024x1024, 1024x1536, and 1536x1024.

GPTProto uses the same key and balance for this model and 200+ others. You can draft with GPT Image 1 Mini, then route selected prompts to a newer model without opening or funding another provider account.

One lifecycle detail should shape your decision. OpenAI deprecated gpt-image-1-mini on June 2, 2026 and scheduled removal from the OpenAI API for December 1, 2026. It remains useful for current budget-sensitive jobs while available, but GPT Image 2 is the safer default for a new application expected to run beyond that date.

GPT Image 1 Mini Specifications

Specification	Value
Provider	OpenAI
GPTProto model string	`gpt-image-1-mini`
Model type	Image generation and image editing
Input modalities	Text and image
Output modality	Image
Standard sizes	1024x1024, 1024x1536, 1536x1024
Quality options	Low, medium, high, auto
Output formats	PNG, JPEG, WebP
GPTProto text input rate	$1.40 per 1M tokens
GPTProto image input rate	$1.75 per 1M tokens
GPTProto image output rate	$5.60 per 1M tokens
Lifecycle	Deprecated by OpenAI; official API removal scheduled for December 1, 2026
Recommended successor	GPT Image 2

GPT Image 1 Mini vs GPT Image 2, Seedream 5 Pro, and Nano Banana 2

GPT Image 1 Mini uses token-based billing, so cost changes with prompt input, image input, output size, and quality. Seedream 5 Pro and Nano Banana 2 use per-image rates on GPTProto, while GPT Image 2 costs more per token but is OpenAI's recommended replacement.

Model	GPTProto billing	Output options	Best fit	Lifecycle
GPT Image 1 Mini	$1.75/M image-input tokens; $5.60/M image-output tokens	1024x1024, 1024x1536, 1536x1024; low/medium/high	Low-cost drafts, prompt testing, and current batch workloads	Deprecated; OpenAI removal scheduled for Dec. 1, 2026
GPT Image 2	$6.40/M image-input tokens; $24.00/M image-output tokens	Flexible sizes up to 4K; low/medium/high	New integrations, higher-fidelity editing, and long-lived products	Current OpenAI replacement
Seedream 5 Pro	$0.0405 per 1K image; $0.081 per 2K image	1K and 2K	Dense layouts, multilingual on-image text, and predictable batch budgets	Current
Nano Banana 2	About $0.04 per generation	1K, 2K, and 4K	Predictable per-image billing and resolution flexibility	Current

GPT Image 1 Mini vs GPT Image 2

Choose GPT Image 1 Mini for inexpensive drafts before its retirement date. Choose GPT Image 2 for a new product, flexible dimensions, higher-fidelity image input, or a workflow that must continue after December 1, 2026. Both use the same GPTProto balance.

GPT Image 1 Mini vs Seedream 5 Pro

GPT Image 1 Mini fits an existing OpenAI-style workflow and token-based budget control. Seedream 5 Pro is better for dense layouts, multilingual copy, or a fixed 1K/2K price. Compare an actual prompt because token and per-image rates are different units.

GPT Image 1 Mini vs Nano Banana 2

Use GPT Image 1 Mini for a low-cost draft tier and familiar GPT Image prompting. Use Nano Banana 2 for a predictable per-call rate or output up to 4K. Test the same prompt and include retries in the comparison.

Switching from OpenAI to GPTProto

For text-to-image generation, keep the model string gpt-image-1-mini and the familiar /v1/images/generations request pattern. Replace the OpenAI base URL with GPTProto's base URL and authenticate with your GPTProto API key. Your prompt and standard size values can stay the same; verify any optional parameter against the GPTProto documentation before moving an existing production request.

One GPTProto key and balance can call GPT Image 1 Mini, GPT Image 2, Seedream, Nano Banana, and other models in the model catalog. During migration, you can test a replacement without opening another provider account.

GPT Image 1 Mini Prompt Guide for Low-Cost Drafts

GPT Image 1 Mini works best when the prompt states the deliverable, composition, subject, lighting, style, and exclusions in that order. Keep text inside the image short, and reserve dense typography or tiny product details for a higher-fidelity model.

Product Concept Draft

Square product concept image for a matte white wireless speaker. Three-quarter front view, centered on a pale gray studio surface, soft shadow to camera left, minimal Scandinavian styling, no logo, no extra objects, clean ecommerce composition.

Social Ad Variation

Portrait 2:3 social ad for a citrus sparkling water. One chilled can in the lower center, orange slices suspended above it, bright cream background, short headline "FRESH START" at the top, bold sans-serif type, generous empty space, no small print.

Image Edit Instruction

Keep the subject, pose, camera angle, crop, and background unchanged. Replace only the jacket with a dark navy linen blazer. Match the original folds, lighting, shadows, and body proportions. Do not change the face, hair, hands, or accessories.

For more budget comparisons, see the cheapest AI image generator guide and the full GPTProto pricing page.

よくある質問

このAIビジョンモデルに関するよくある質問

What is the GPT Image 1 Mini API?

The GPT Image 1 Mini API provides programmatic access to OpenAI's cost-efficient image model. It accepts text and image input and returns image output. On GPTProto, use the model string gpt-image-1-mini with one GPTProto API key.

How much does GPT Image 1 Mini cost on GPTProto?

GPTProto lists text input at $1.40, image input at $1.75, and image output at $5.60 per 1M tokens, 30% below OpenAI's corresponding rates. Final cost varies with input, dimensions, and quality; use the live calculator above for batch estimates.

Does GPT Image 1 Mini support image editing?

Yes. OpenAI defines it as a multimodal model that accepts both text and image input and produces image output. Use an image-edit request when you need to transform or revise an existing image rather than generate one from text alone.

Which sizes and quality levels does GPT Image 1 Mini support?

The standard output sizes are 1024x1024, 1024x1536, and 1536x1024. Supported quality settings are low, medium, high, and auto. Larger sizes and higher quality use more output tokens and therefore cost more.

Is GPT Image 1 Mini still supported?

It is available now, but OpenAI has deprecated the model and scheduled removal from the OpenAI API on December 1, 2026. GPTProto availability after that date should be confirmed before committing to a long-term deployment. OpenAI recommends GPT Image 2 as the replacement.

Is GPT Image 1 Mini or GPT Image 2 better?

GPT Image 1 Mini is the budget choice for current drafts and prompt tests. GPT Image 2 is the better default for new applications because it supports flexible sizes, processes image input at high fidelity, and is the recommended successor for workloads that must continue after December 1, 2026.

Can I get a free GPT Image 1 Mini API key?

GPTProto uses pay-as-you-go account balance rather than a permanent free API tier. Create one API key, add balance, and use it across supported models. Check the current signup or promotional offer before claiming any free credit because those offers can change.