OpenAI

llms.txt

GPT Image 1.5

Name: GPT Image 1.5
Brand: OpenAI
Price: 0.009 USD
Availability: InStock

AI Image Editing Model

Mode:

All Text to Image Image Edit

Image #7 Text-to-Image #3 Image-to-Image

OpenAI's state-of-the-art image generation model with better instruction following and adherence to prompts

GPT Image 1.5 generated image of A hyper-realistic close-up of a weathered craftsman's workbench, featuring a ...

GPT Image 1.5 generated image of A hyper-realistic cinematic close-up of an elderly watchmaker’s workspace, su...

Try Model Benchmark examples

1536 x 1536

Max Resolution

Supported Modes

Text to Image Image Edit

Active

Details

Model ID

gpt-image-1.5

Creator

OpenAI

Family

gpt-image

Released

December 2025

Max Input Images

Ready to integrate?

Access gpt-image-1.5 via our unified API.

Create Account

Available at 3 providers

Starting from

$0.0090 /image via fal.ai, OpenAI · +1 more

By quality

Low

~$0.010

Medium

~$0.039

High

~$0.134

Prices shown are in USD

See all providers

Provider Performance

Fastest generation through replicate at 34,702ms median latency with 94.1% success rate.

Aggregated from real API requests over the last 30 days.

Generation Time

replicate

p95: 42,566ms

openai

p95: 58,033ms

fal

p95: 87,593ms

Success Rate

replicate

94.1%

16 / 17 requests

openai

75.9%

308 / 406 requests

fal

100.0%

60 / 60 requests

Time to First Byte

replicate

35,276ms

p95: 38,198ms

openai

38,727ms

p95: 57,434ms

fal

50,112ms

p95: 87,205ms

Provider Rankings

#	Provider	p50 Gen Time	p95 Gen Time	Success Rate	TTFB (p50)
1	replicate	34,702ms	42,566ms	94.1%	35,276ms
2	openai	39,330ms	58,033ms	75.9%	38,727ms
3	fal	49,823ms	87,593ms	100.0%	50,112ms

Data updated every 15 minutes. Based on all API requests through Lumenfall over the last 30 days.

Providers & Pricing (3)

GPT Image 1.5 is available from 3 providers, with per-image pricing starting at $0.009 through fal.ai.

fal.ai

Image Edit

fal/gpt-image-1.5-edit

Provider Model ID: fal-ai/gpt-image-1.5/edit

Output

Image 1024x1024, High

$0.133 per image

Image 1024x1024, Low

$0.0090 per image

Image 1024x1024, Medium

$0.034 per image

Image 1024x1536, High

$0.200 per image

Image 1024x1536, Low

$0.013 per image

Image 1024x1536, Medium

$0.051 per image

Image 1536x1024, High

$0.199 per image

Image 1536x1024, Low

$0.013 per image

Image 1536x1024, Medium

$0.050 per image

Pricing Notes (1)

• Per-image prices vary by quality (low/medium/high) and size

View official pricing • As of Jan 27, 2026

Replicate

Text to Image Image Edit

replicate/gpt-image-1.5

Provider Model ID: openai/gpt-image-1.5

Output

Image High

$0.136 per image

Image Low

$0.013 per image

Image Medium

$0.050 per image

Pricing Notes (1)

• Per-image prices vary by quality level

View official pricing • As of Jan 27, 2026

OpenAI

Text to Image Image Edit

openai/gpt-image-1.5

Provider Model ID: gpt-image-1.5

Input

Token text

$5.00 per 1M

Token text Cached

$1.25 per 1M

Token image

$8.00 per 1M

Token image Cached

$2.00 per 1M

Output

Token text

$10.00 per 1M

Token image

$32.00 per 1M

Image 1024x1024, High

$0.133 per image

Image 1024x1024, Low

$0.0090 per image

Image 1024x1024, Medium

$0.034 per image

Image 1024x1536, High

$0.200 per image

Image 1024x1536, Low

$0.013 per image

Image 1024x1536, Medium

$0.050 per image

Image 1536x1024, High

$0.200 per image

Image 1536x1024, Low

$0.013 per image

Image 1536x1024, Medium

$0.050 per image

Pricing Notes (3)

• Pricing available in both token-based and per-image formats
• Per-image prices vary by quality (low/medium/high) and size
• Supports up to 10 input images; first 5 preserved with higher fidelity

View official pricing

GPT Image 1.5 API OpenAI-compatible

Developers can integrate GPT Image 1.5 through the Lumenfall OpenAI-compatible API to automate text-to-image generation and complex image editing workflows. This single endpoint provides programmatic access to OpenAI's state-of-the-art vision capabilities for scalable media production.

Base URL

https://api.lumenfall.ai/openai/v1

Model

gpt-image-1.5

Code Examples

Image Edit

/v1/images/edits

curl -X POST \

  https://api.lumenfall.ai/openai/v1/images/edits \

  -H "Authorization: Bearer $LUMENFALL_API_KEY" \

  -F "model=gpt-image-1.5" \

  -F "[email protected]" \

  -F "prompt=Add a starry night sky to this image" \

  -F "size=1024x1024"

# Response:

# { "created": 1234567890, "data": [{ "url": "https://...", "revised_prompt": "..." }] }

import OpenAI from 'openai';

import fs from 'fs';

const client = new OpenAI({

  apiKey: 'YOUR_API_KEY',

  baseURL: 'https://api.lumenfall.ai/openai/v1'

});

const response = await client.images.edit({

  model: 'gpt-image-1.5',

  image: fs.createReadStream('source.png'),

  prompt: 'Add a starry night sky to this image',

  size: '1024x1024'

});

// { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }

console.log(response.data[0].url);

from openai import OpenAI

client = OpenAI(

    api_key="YOUR_API_KEY",

    base_url="https://api.lumenfall.ai/openai/v1"

response = client.images.edit(

    model="gpt-image-1.5",

    image=open("source.png", "rb"),

    prompt="Add a starry night sky to this image",

    size="1024x1024"

# { created: 1234567890, data: [{ url: "https://...", revised_prompt: "..." }] }

print(response.data[0].url)

Parameter Reference

Required Supported Not available

Core Parameters

Parameter	Type	Description	Modes
`prompt`	string	Required. Text prompt for image generation	T2I Edit

Size & Layout

Parameter	Type	Description	Modes
`size`	string	Image dimensions as WxH pixels (e.g. "1024x1024") or aspect ratio (e.g. "16:9") `1254x836` `836x1254` `1024x1024` `1536x1024` `1024x1536` WxH determines both shape and scale (aspect_ratio and resolution are ignored when size is provided). W:H format is equivalent to aspect_ratio.	T2I Edit
`aspect_ratio`	string	Aspect ratio of the output image (e.g. "16:9", "1:1") `2:3` `1:1` `3:2` Controls shape independently of scale. Use with resolution to control both. If size is also provided, size takes precedence. Any ratio is accepted and mapped to the nearest supported value.	T2I Edit
`resolution`	string	Output resolution tier (e.g. "1K", "4K") `1K` Controls scale independently of shape. Higher tiers produce larger images and cost more. If size is also provided, size takes precedence for scale. Any tier is accepted and mapped to the nearest supported value.	T2I Edit

Output	`size`		`aspect_ratio` + `resolution`
Flexible
Custom 1024–1536px per side	`"WxH"`	—	Any pixel dimensions within model constraints

1K 5 sizes

Output	`size`		`aspect_ratio` + `resolution`
1024 × 1536	`"1024x1536"`	or	`"2:3"` + `"1K"`
1536 × 1024	`"1536x1024"`	or	`"3:2"` + `"1K"`
1024 × 1024	`"1024x1024"`	or	`"1:1"` + `"1K"`
836 × 1254	`"836x1254"`	or	`"2:3"` + `"1K"`
1254 × 836	`"1254x836"`	or	`"3:2"` + `"1K"`

How these parameters work

size

Exact pixel dimensions

"1920x1080"

aspect_ratio

Shape only, default scale

"16:9"

resolution

Scale tier, preserves shape

"1K"

Priority when combined

size › aspect_ratio + resolution › aspect_ratio › resolution

size is most specific and always wins. aspect_ratio and resolution control shape and scale independently.

How matching works

Shape matching – we pick the closest supported ratio. Ask for 7:1 on a model with 4:1 and 8:1, you get 8:1.

Scale matching – providers use different tier formats: K tiers (0.5K 1K 2K 4K) or megapixel tiers (0.25 1). If the exact tier isn't available, you get the nearest one.

Dimension clamping – if a model has pixel limits, we clamp dimensions to fit and keep the aspect ratio intact.

Media Inputs

Parameter	Type	Description	Modes
`image`	file	Required. Input image(s) to edit Supports PNG, JPEG, WebP. Up to 10 images per request.	T2I Edit

Output & Format

Parameter	Type	Description	Modes
`response_format`	string	How to return the image `url` `b64_json` Default: `"url"`	T2I Edit
`output_format`	string	Output image format `png` `jpeg` `gif` `webp` `avif` Gateway converts to requested format if provider doesn't support it natively.	T2I Edit
`output_compression` replicate	integer	Compression level (0-100%) Default: `90`	T2I Edit
`n`	integer	Number of images to generate Default: `1` Gateway generates multiple images in parallel even if provider only supports 1.	T2I Edit

Additional Parameters

Provider-specific passthrough fields, available only when the request is routed to the listed provider.

Parameter	Type	Description	Modes
Universal
`background`	string	Set whether the background is transparent or opaque or choose automatically `opaque` `transparent`	T2I Edit
`input_fidelity`	string	Control how much effort the model will exert to match the style and features, especially facial features, of input images `high` `low`	T2I Edit
`moderation`	string	Content moderation level `low`	T2I Edit
fal
`mask_image_url`	string	The URL of the mask image to use for the generation. This indicates what part of the image to edit.	T2I Edit
`sync_mode`	boolean	If `True`, the media will be returned as a data URI and the output data won't be available in the request history.	T2I Edit
openai
`partial_images`	integer	Number of partial images to generate in streaming responses.	T2I Edit
`stream`	boolean	Whether to stream image generation or edit events as they become available.	T2I Edit
`user`	string	Stable end-user identifier used by OpenAI abuse monitoring.	T2I Edit
replicate
`openai_api_key`	string	Your OpenAI API key (optional - uses proxy if not provided)	T2I Edit
`user_id`	string	An optional unique identifier representing your end-user. This helps OpenAI monitor and detect abuse.	T2I Edit

Parameter Normalization

How we handle parameters across different providers

Not every provider speaks the same language. When you send a parameter, we handle it in one of four ways depending on what the model supports:

Behavior	What happens	Example
`passthrough`	Sent as-is to the provider	style, quality
`renamed`	Same value, mapped to the field name the provider expects	prompt
`converted`	Transformed to the provider's native format	size
`emulated`	Works even if the provider has no concept of it	n, response_format

Parameters we don't recognize pass straight through to the upstream API, so provider-specific options still work.

Full API Reference Authentication, endpoints, and more

GPT Image 1.5 Benchmarks

GPT Image 1.5 holds an Elo rating of 1285 in text-to-image generation, ranking #2 globally, and maintains a #3 rank for image editing with an Elo of 1230. These scores reflect its superior instruction following and competitive performance across major open-vision leaderboards.

Lumenfall Arena

Text-to-Image

1267 Elo

Lumenfall Arena

Image Editing

1229 Elo

Image Editing Landscape

Elo vs Cost

1 model without pricing omitted

Elo vs Speed

7 models waiting for enough speed data

Competition Results

Image Editing

Photorealism

View leaderboard

Man and Car in California

13 models

Image Editing

Source

Edit instruction

“Make a photo of the man driving the car down the California coastline”

GPT Image 1.5 edited result for Golden Hour Stroll

Original image before GPT Image 1.5 editing

Before After

Golden Hour Stroll

13 models

Image Editing

Edit instruction

“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”

GPT Image 1.5 edited result for Bald man challenge

Before After

#11

Bald man challenge

15 models

Image Editing

Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

GPT Image 1.5 edited result for Neutral Expression to Genuine Smile

Before After

Neutral Expression to Genuine Smile

14 models

Image Editing

Edit instruction

{
  "action": "image_edit",
  "reference": "uploaded neutral portrait",
  "change": "Warm genuine Duchenne smile: lips curved up, slight natural teeth, soft eye crinkles, subtle cheek raise",
  "details": "Realistic smiling skin (dimples if present, soft cheek shadows), slightly brighter eyes; keep exact eye shape/color/iris",
  "preserve_exact": "Face identity/structure, eyes/nose/lips/eyebrows, hair, skin texture/pores/freckles, makeup, clothing, head pose, background, lighting, shadows, framing",
  "no_changes": "No face shape change, no new features, no gaze shift, no hair/clothing/lighting/background edits",
  "style": "Ultra-photorealistic 8K portrait, sharp face focus, natural soft lighting, realistic skin glow"
}

3 attempts – showing best result

GPT Image 1.5 edited result for Night Sky Transformation

Before After

Night Sky Transformation

16 models

Image Editing

Edit instruction

“Change the scene to night: a deep, dark sky with subtle, glistening stars visible behind the mountain.”

3 attempts – showing best result

Image Editing

Anime

View leaderboard

#11

Studio Ghibli Anime Style

14 models

Image Editing

Source

Edit instruction

“Transform this photo into a Studio Ghibli–inspired illustration. Use soft pastel colors, hand-painted textures, gentle lighting, dreamy backgrounds, and a warm, nostalgic mood”

Image Editing

Portrait

View leaderboard

Before After

#11

Bald man challenge

15 models

Image Editing

Edit instruction

“Give the person a full, thick head of natural hair with realistic texture, density, and a natural hairline. Preserve facial features and lighting.”

Before After

Neutral Expression to Genuine Smile

14 models

Image Editing

Edit instruction

{
  "action": "image_edit",
  "reference": "uploaded neutral portrait",
  "change": "Warm genuine Duchenne smile: lips curved up, slight natural teeth, soft eye crinkles, subtle cheek raise",
  "details": "Realistic smiling skin (dimples if present, soft cheek shadows), slightly brighter eyes; keep exact eye shape/color/iris",
  "preserve_exact": "Face identity/structure, eyes/nose/lips/eyebrows, hair, skin texture/pores/freckles, makeup, clothing, head pose, background, lighting, shadows, framing",
  "no_changes": "No face shape change, no new features, no gaze shift, no hair/clothing/lighting/background edits",
  "style": "Ultra-photorealistic 8K portrait, sharp face focus, natural soft lighting, realistic skin glow"
}

3 attempts – showing best result

Uncategorized

Over-the-top cartoon caricature

13 models

Image Editing

Source

Edit instruction

“Create a caricature of me and my job. Make it exaggerated and humorous, incorporating my profession as a tv show anchor and my love for dogs and hockey.”

Pose & Character Mashup

14 models

Image Editing

Source

Edit instruction

“Use Image 1 as the exact pose reference and Image 2 as the character reference. Recreate the person/character from Image 2 in the exact dynamic pose and body position from Image 1. Keep the exact face, hair, clothing style/details, and expression from Image 2. Match the lighting and environment of Image 1. The final image must show the character from Image 2 performing the precise action/pose from Image 1 with perfect anatomy and natural integration.”

Top Matchups

See how GPT Image 1.5 performs head-to-head against other AI models, ranked by community votes in blind comparisons.

vs Grok Imagine Image

Challenge: Modern Clean Menu

50% W · 50% L

vs Nano Banana 2

Challenge: Fantasy Warrior

0% W · 50% L · 50% T

vs ImagineArt 1.5 (Preview)

Challenge: Geometric Composition

50% W · 50% L

vs Wan 2.6

Challenge: Golden Hour Stroll

50% W · 50% L

vs Nano Banana

Challenge: Bald man challenge

0% W · 100% L

GPT Image 1.5 is best for

See all Use Cases

GPT Image 1.5 leads the Portrait category with a first-place rank and 84.6% win rate, while also ranking #3 for Text Rendering and Product/Commercial usage. It performs significantly weaker in Anime generation, where it currently ranks #13 with a 26.3% win rate.

#3 of 21

Photorealism

Edit · 58.1% win rate

Gallery

View all 8 images

GPT Image 1.5 FAQ

How much does GPT Image 1.5 cost?

GPT Image 1.5 starts at $0.009 per image through Lumenfall. Pricing varies by provider. Lumenfall does not add any markup to provider pricing.

How do I use GPT Image 1.5 via API?

You can use GPT Image 1.5 through Lumenfall's OpenAI-compatible API. Send requests to the unified endpoint with model ID "gpt-image-1.5". Code examples are available in Python, JavaScript, and cURL.

Which providers offer GPT Image 1.5?

GPT Image 1.5 is available through Replicate, OpenAI, and fal.ai on Lumenfall. Lumenfall automatically routes requests to the best available provider.

What is the maximum resolution for GPT Image 1.5?

GPT Image 1.5 supports images up to 1536x1536 resolution.

Overview

GPT Image 1.5 is OpenAI’s latest flagship image generation model, designed to transform complex text descriptions into high-fidelity visual assets. It represents a significant iteration in the GPT-image family, focusing on narrowing the gap between user intent and generated output. The model is distinctive for its high level of steerability, allowing users to define specific spatial arrangements and intricate details that previous iterations often elided.

Strengths

Complex Instruction Following: The model excels at parsing long, multi-part prompts, ensuring that every requested element is present in the final composition without losing track of secondary details.
Spatial and Relational Accuracy: It maintains high consistency when placing objects in specific locations (e.g., “in the bottom left corner”) or defining relationships between subjects (e.g., “leaning against” or “half-obscured by”).
Text Rendering Accuracy: GPT Image 1.5 shows marked improvement in rendering legible, correctly spelled text within images, making it suitable for graphic design mockups and signage.
Diverse Aspect Ratios: Unlike earlier generative models restricted to square outputs, this model natively supports a wide range of aspect ratios while maintaining structural integrity and avoiding anatomical distortion.

Limitations

Photorealistic Nuance: While highly capable, it may still struggle with specific “uncanny valley” effects in human skin textures or micro-expressions compared to specialized diffusion models tuned specifically for photography.
Prompt Literalism: Because the model prioritizes strict adherence to instructions, it can occasionally lack the “artistic flair” or unexpected creativity found in models that interpret prompts more loosely.
Inference Latency: Given the complexity of the architecture required to achieve high instruction following, generation times may be slightly higher than smaller, distilled latent diffusion models.

Technical Background

GPT Image 1.5 is built upon a transformer-based diffusion architecture, leveraging OpenAI’s advancements in large-scale multimodal pre-training. By utilizing a sophisticated text encoder similar to those found in the GPT-4 family, the model can internalize nuanced semantic meanings before translating them into the visual latent space. This architecture enables the model to treat image generation as a sequence-informed task, improving the alignment between the text tokens and the resulting pixels.

Best For

GPT Image 1.5 is ideal for professional workflows including advertising concept art, architectural visualization, and detailed character design where precision is non-negotiable. Its ability to follow strict formatting makes it a strong candidate for automated content pipelines and social media asset generation.

You can experiment with GPT Image 1.5 and compare its outputs with other top-tier models through the Lumenfall playground or integrate it into your production environment using the Lumenfall unified API.

Try GPT Image 1.5 in Playground

Generate images with custom prompts — no API key needed.

Open Playground