Price directory

Image generation API pricing, per image or in credits.

33 image models — per image, per megapixel, or in credits; every price dated and sourced.

Latest events

GPT-5.6 Terra cut 20%

Jul 31, 2026

Terra's blended rate now works out to $14 per million tokens ($2 in / $12 out), which puts it below Anthropic's mid-tier Claude Sonnet 4.6, priced at $3 per million input tokens and $15 per million output tokens — so any routing logic picking models on cost-per-token alone now favors Terra over Sonnet by a real marg...

Claude Opus 5: Fable perf, half price Jul 24, 2026

Kimi K3: Opus-class at $3/$15 Jul 16, 2026

Inkling opens Apache 2.0 frontier Jul 15, 2026

DeepSeek V4 adds peak-valley pricing Jun 29, 2026

View timeline →

Price table

33 models of 33

Model	Provider	Pricing	Released
Stable Diffusion 3.5 Large Stability AI Image generation Stability AI's 8B-parameter text-to-image model, on the hosted API and as open weights. —/—I/O View model	Stability AI	6.5 credits ≈ $0.065 Credits as of Jul 2026	—
B Seedream 4.5 ByteDance Image generation ByteDance's text-to-image model, served through BytePlus ModelArk. —/—I/O View model	ByteDance	$0.04 / image Per image as of Jul 2026	—
R Reve Reve Image generation Reve's text-to-image model, in beta, via its API. —/—I/O View model	Reve	5 credits ≈ $0.0067 Credits as of Jul 2026	—
R Recraft V3 Recraft Image generation Recraft's text-to-image model with raster and vector output, via its API. —/—I/O View model	Recraft	$0.04 raster · $0.08 vector Per image as of Jul 2026	—
Qwen-Image Alibaba Image generation Alibaba's open-weight text-to-image model, on Model Studio and as open weights. —/—I/O View model	Alibaba	Pricing not published	—
L Photon Luma AI Image generation Luma AI's text-to-image model, via its API. —/—I/O View model	Luma AI	$0.015 / image Per image as of Jul 2026	—
A Nova Canvas Amazon Image generation Amazon's text-to-image model, available through Amazon Bedrock. —/—I/O View model	Amazon	$0.04–$0.08 / image Per image as of Jul 2026	—
Nano Banana Pro (Gemini 3 Pro Image) Google Image generation Nano Banana Pro is Google's image-and-text model built on the Gemini 3 Pro architecture, positioned as an older generation compared to the later Nano Banana 2 series released in 2026-06. $2.00/$12.00I/O 65K ctx Jun 2026 View model	Google	$2.00/$12.00	Jun 2026
Nano Banana Pro Google Image generation Google's current flagship image model (Gemini 3 Pro Image), via the Gemini API. —/—I/O View model	Google	$0.134–$0.24 / image Token-based as of Jul 2026	—
Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) Google Image generation Google's lightweight image-and-text model built on the Gemini 3.1 Flash architecture, designed for efficient multimodal tasks with a 65K token context window. $0.25/$1.50I/O 65K ctx Jun 2026 View model	Google	$0.25/$1.50	Jun 2026
Nano Banana 2 (Gemini 3.1 Flash Image) Google Image generation Google's lightweight multimodal model that processes text and images, built on the Gemini 3.1 Flash architecture with a 131K token context window. $0.5/$3.00I/O 131K ctx Jun 2026 View model	Google	$0.5/$3.00	Jun 2026
Nano Banana 2 Google Image generation Google's Gemini 3.1 Flash Image model, via the Gemini API. —/—I/O View model	Google	$0.045–$0.151 / image Token-based as of Jul 2026	—
Nano Banana (Gemini 2.5 Flash Image) Google Image generation Gemini 2.5 Flash Image is Google's multimodal model from the Gemini 2.5 family, handling text and image inputs with native thinking capabilities. $0.3/$2.50I/O 32K ctx Oct 2025 View model	Google	$0.3/$2.50	Oct 2025
Muse Image Meta Image generation Meta Superintelligence Labs' first image generation model — follows instructions, edits with precision, and composes from multiple references. Available in the Meta AI app, Instagram Stories (US), and WhatsApp (limited countries). —/—I/O View model	Meta	Pricing not published	—
L Leonardo AI Leonardo AI Image generation Leonardo AI's text-to-image platform, via its API. —/—I/O View model	Leonardo AI	Varies by model & resolution Credits as of Jul 2026	—
Imagen 4 Google legacy Image generation Google's text-to-image model, via the Gemini API and Vertex AI. —/—I/O View model	Google	$0.02–$0.06 / image Per image as of Jul 2026	—
I Ideogram 3.0 Ideogram Image generation Ideogram's text-to-image model, built to render legible text and typography within the image, via its API. —/—I/O View model	Ideogram	$0.03–$0.09 / image Per image as of Jul 2026	—
Grok Imagine Image xAI Image generation xAI's text-to-image model, via the xAI API. —/—I/O View model	xAI	$0.02–$0.05 / image Per image as of Jul 2026	—
GPT-5.4 Image 2 OpenAI Image generation GPT-5.4 Image 2 is OpenAI's image-specialized model from the GPT-5.4 series with a 272k token context window, supporting multimodal inputs for image analysis and understanding. $8.00/$15.00I/O 272K ctx Apr 2026 View model	OpenAI	$8.00/$15.00	Apr 2026
GPT-5 Image Mini OpenAI Image generation A lightweight image generation and understanding model from OpenAI with a 400k token context window, positioned as a faster alternative to GPT-5 Image. $2.50/$2.00I/O 400K ctx Oct 2025 View model	OpenAI	$2.50/$2.00	Oct 2025
GPT-5 Image OpenAI Image generation OpenAI's multimodal model that handles both image generation and understanding tasks with a 400k token context window. $10.00/$10.00I/O 400K ctx Oct 2025 View model	OpenAI	$10.00/$10.00	Oct 2025
GPT Image 2 OpenAI Image generation OpenAI's current flagship image model, exposed through the Images and Responses APIs. —/—I/O View model	OpenAI	$0.006–$0.211 / image Token-based as of Jul 2026	—
GPT Image 1.5 OpenAI legacy Image generation Previous-generation OpenAI image model, still callable while it winds down. —/—I/O View model	OpenAI	$0.009–$0.133 / image Token-based as of Jul 2026	—
GPT Image 1 Mini OpenAI legacy Image generation Smaller, cheaper OpenAI image model from the GPT Image 1 line. —/—I/O View model	OpenAI	$0.005–$0.036 / image Token-based as of Jul 2026	—
GPT Image 1 OpenAI legacy Image generation OpenAI's first natively multimodal image model, now deprecated. —/—I/O View model	OpenAI	$0.011–$0.167 / image Token-based as of Jul 2026	—
Gemini 2.5 Flash Image Google Image generation Google's "Nano Banana" image model, available in preview through the Gemini API. —/—I/O View model	Google	≈ $0.039 / image Token-based as of Jul 2026	—
B FLUX1.1 [pro] Black Forest Labs Image generation Black Forest Labs' text-to-image model, served through their API and inference partners. —/—I/O View model	Black Forest Labs	$0.04 / image Per image as of Jul 2026	—
B FLUX.2 [pro] Black Forest Labs Image generation Black Forest Labs' current flagship text-to-image model. —/—I/O View model	Black Forest Labs	from $0.03 / MP Per megapixel as of Jul 2026	—
B FLUX.1 [pro] Black Forest Labs Image generation Black Forest Labs' flagship text-to-image model, served through their API and several inference providers. Billed per generated image. —/—I/O Aug 2024 View model	Black Forest Labs	Not yet tracked	Aug 2024
A Firefly Adobe Image generation Adobe's text-to-image model, offered through Firefly Services for enterprises. —/—I/O View model	Adobe	~10 credits / image Credits as of Jul 2026	—
B Bria Bria AI Image generation Bria AI's commercially-licensed text-to-image models, via its API. —/—I/O View model	Bria AI	$0.02–$0.03 / image Per image as of Jul 2026	—
O Auto Router (Beta) OpenRouter Image generation A routing meta-model that classifies incoming requests and directs them to downstream models chosen for the specific task. —/—I/O 2M ctx Jul 2026 View model	OpenRouter	Not yet tracked	Jul 2026
O Auto Router OpenRouter Image generation A routing meta-model that classifies incoming requests and directs them to downstream models chosen for the specific task. —/—I/O 2M ctx Nov 2023 View model	OpenRouter	Not yet tracked	Nov 2023