impulse

GPT Image 2

Frontier labImageRetiring 10 Feb 2027
API id gpt-image-2Released 21 Apr 2026 (Apr 2026)
OpenAI retires it on 10 Feb 2027 (in 84 days). 4 hosts keep serving it.
Input / 1M
$5.00
at OpenAI
Output / 1M
-
at OpenAI
Context window
-
tokens
Max output
-
tokens per reply
Hosts
5
25 listings
Released
Apr 2026
21 Apr 2026

Pricing

List prices at OpenAI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$5.00
Cached input
75% off input
$1.25
Batch
50% off
$2.50

What it costs a month

At OpenAI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$1,163
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$1,500
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$2,850
  • Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$6,000
Use your own numbers

Where to run it

5 hosts serve GPT Image 2. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
OpenAILab API
gpt-image-2
$5.00 / -
Retires 10 Feb 2027
gpt-image-2
$5.00 / -
Retires 21 Oct 2027
gpt-image-2
$5.00 / -
Retires 21 Oct 2027
databricks-gpt-image-2
$5.00 / -
falInference
gpt-image-2
-
Show 20 more listings: regions, snapshots and variants
OpenAI
2026-04-21
gpt-image-2-2026-04-21
$5.00 / -
Retires 10 Feb 2027
Azure OpenAI
2026-04-21
gpt-image-2-2026-04-21
$5.00 / -
Retires 21 Oct 2027
fal
low · 1024×768
gpt-image-2
-
low · 1024×1024
gpt-image-2
-
low · 1024×1536
gpt-image-2
-
low · 1920×1080
gpt-image-2
-
low · 2560×1440
gpt-image-2
-
low · 3840×2160
gpt-image-2
-
medium · 1024×768
gpt-image-2
-
medium · 1024×1024
gpt-image-2
-
medium · 1024×1536
gpt-image-2
-
medium · 1920×1080
gpt-image-2
-
medium · 2560×1440
gpt-image-2
-
medium · 3840×2160
gpt-image-2
-
high · 1024×768
gpt-image-2
-
high · 1024×1024
gpt-image-2
-
high · 1024×1536
gpt-image-2
-
high · 1920×1080
gpt-image-2
-
high · 2560×1440
gpt-image-2
-
high · 3840×2160
gpt-image-2
-

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImagesPDFs
Produces
Images
Features
PDF input

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than OpenAI.

Smaller OpenAI models

No cheaper OpenAI model keeps its key capabilities.

Other frontier labs

Open-weight picks

No cheaper open-weight match.

Model line

Newer and older versions, at each lab's own price.

  1. Apr 2026GPT Image 2Latest$5.00 / -
  2. Dec 2025GPT Image 1.5$5.00 / $10.00
  3. UndatedGPT Image 1$5.00 / -

Call it

Using OpenAI's API.

Python
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY

response = client.responses.create(model="gpt-image-2", input="Hello")
print(response.output_text)

Quick answers

How much does GPT Image 2 cost?

On OpenAI, GPT Image 2 costs $5.00 per million input tokens. Cached input costs $1.25 per million tokens.

Which APIs offer GPT Image 2?

GPT Image 2 runs on 5 APIs: OpenAI, Azure OpenAI, Azure AI Foundry, Databricks and fal.

Is GPT Image 2 being retired?

OpenAI retires it on 10 Feb 2027; Azure AI Foundry and Azure OpenAI on 21 Oct 2027. 2 hosts keep serving it.

Is there a newer version of GPT Image 2?

No. GPT Image 2 is the newest model in its line.

What are cheaper alternatives to GPT Image 2?

Models that cost less and keep most of what it can do: Gemini 3 Pro Image on Google AI Studio (10% less) and Gemini Nano Banana 2.1 on Google AI Studio (40% less).

When was GPT Image 2 released?

OpenAI released it on 21 Apr 2026.

What is GPT Image 2's API model id?

Use gpt-image-2 with OpenAI. Other hosts use their own ids: databricks-gpt-image-2 on Databricks.

Sources

Raw catalog entry for gpt-image-2 on OpenAI
{
  "cache_read_input_image_token_cost": 0.000002,
  "cache_read_input_image_token_cost_batches": 0.000001,
  "cache_read_input_token_cost": 0.00000125,
  "cache_read_input_token_cost_batches": 6.25e-7,
  "deprecation_date": "2027-02-10",
  "deprecation_source": "https://developers.openai.com/api/docs/deprecations",
  "input_cost_per_token": 0.000005,
  "litellm_provider": "openai",
  "mode": "image_generation",
  "input_cost_per_image_token": 0.000008,
  "input_cost_per_image_token_batches": 0.000004,
  "input_cost_per_token_batches": 0.0000025,
  "output_cost_per_image_token": 0.00003,
  "output_cost_per_image_token_batches": 0.000015,
  "source": "https://developers.openai.com/api/docs/pricing",
  "successor_model": "gpt-image-2.6-sunburst",
  "supported_endpoints": [
    "/v1/images/generations",
    "/v1/images/edits"
  ],
  "supports_vision": true,
  "supports_pdf_input": true
}