AI image generation API pricing
API prices for 41 AI image models from 15 companies, each as the vendor lists it, with the top resolution and whether the model also edits existing images.
| Model | Company | API price | Resolution | Image editing | Released |
|---|---|---|---|---|---|
| FLUX 3 Image Klein | Black Forest Labs | $0.008 (768 × 768), $0.009 (1K), $0.011 (1.5K), $0.014 (2K), $0.036 (4K) | Up to 4K | Yes - up to 10 reference images | Oct 2026 |
| Gemini Nano Banana 2.1 | $0.0336 (1K, 1024x1024), $0.0504 (2K), $0.113 (4K); Batch half price | 4K (4096x4096) | — | Oct 2026 | |
| FLUX 3 Image | Black Forest Labs | $0.041 (768 × 768), $0.048 (1K), $0.07 (1.5K), $0.10 (2K), $0.607 (4K) | 4K (about 16 megapixels), 15 aspect ratios | Yes - up to 10 reference images | Oct 2026 |
| MAI Image 2.6 | Microsoft AI | $5 text / $8 image input, $38 image output per 1M tokens (Microsoft Foundry, effective 1 September 2026) | Up to 2,359,296 total pixels (equivalent to 1536×1536) | Yes — image edits API (JPEG or PNG in, PNG out) | Jul 2026 |
| MAI Image 2.6 Flash | Microsoft AI | $1.75 text / $2.50 image input, $19 output per 1M tokens | Up to 2,359,296 total pixels (equivalent to 1536×1536) | Yes — image edits API (JPEG or PNG in, PNG out) | Jul 2026 |
| Midjourney V8.2 | Midjourney | No per-image price - subscription + fast-GPU-minute billing ($4/hr extra Fast GPU time) | 2048x2048px HD (native V8.2, default 1:1 ratio); 1024x1024px SD; further upscaling via third-party tools | Yes (Image Prompts) | Jul 2026 |
| Muse Image | Meta | $0.01 (muse-image-1.0, Meta Model API) | — | Yes — edits and composes with multiple reference images, refined turn by turn | Jul 2026 |
| Qwen Image 3.0 Pro | Alibaba | $0.04 per output image, $0.003 per input image (1K, International) | 2048×2048 (512×512 to 2048×2048) | — | Jul 2026 |
| Seedream 5.0 Pro | ByteDance | $0.045 (1.5K or lower), $0.09 (above 1.5K); layer decomposition $0.0225 / $0.045 | 2K (1K, 1.5K or 2K) | Yes — single or multiple reference images | Jun 2026 |
| MAI Image 2.5 Pro | Microsoft AI | $5 text / $8 image input, $106 output per 1M tokens | About 1 megapixel (up to 1,048,576 total pixels; each side at least 768) | Yes — image edits API (JPEG or PNG in, PNG out) | Jun 2026 |
| Ideogram 4.0 | Ideogram | $0.03 (Turbo), $0.06 (Default), $0.10 (Quality) | 2,048 pixels; transparent-background images up to 8K through the API | Yes — Remix | Jun 2026 |
| MAI Image 2.5 | Microsoft AI | $5 text / $8 image input, $47 image output per 1M tokens (Microsoft Foundry) | About 1 megapixel (width × height up to 1,048,576; each side at least 768) | Yes — image edits API (JPEG or PNG in, PNG out) | Jun 2026 |
| MAI Image 2.5 Flash | Microsoft AI | $1.75 text / $1.75 image input, $19.50 output per 1M tokens | About 1 megapixel (up to 1,048,576 total pixels; each side at least 768) | Yes — image edits API (JPEG or PNG in, PNG out) | Jun 2026 |
| Grok Imagine Image 2.0 | SpaceXSI | $0.04 | — | Yes — edits from up to 5 source images | Jun 2026 |
| Uni 1.1 Max | Luma AI | $0.1000 (2K text to image), $0.1030 (image edit) — Luma API | 2,048 pixels (2K) | Yes — image edit, with up to 8 image references | Jun 2026 |
| Recraft V4.1 | Recraft | $0.035 raster, $0.08 vector (SVG) | — | Yes — raster and vector | Jun 2026 |
| Gemini 3.1 Flash Image | $0.045 (0.5K), $0.067 (1K); batch $0.022 / $0.034 | 4K (512px, 1K, 2K, 4K) | Yes — up to 14 reference images | May 2026 | |
| Gemini 3.1 Flash Image (Nano Banana 2) | Standard: $0.045 (0.5K/512px), $0.067 (1K), $0.101 (2K), $0.151 (4K); Batch API (50% off): $0.022/$0.034/$0.050/$0.076 | Up to 4096x4096px (4K, 1:1 ratio); other ratios scale within the same per-tier token budget, e.g. 12288x1536 at 8:1. Also 512x512 (0.5K/512px), 1024x1024 (1K, default), 2048x2048 (2K) | Yes (text-and-image-to-image editing, up to 14 total reference images: up to 10 high-fidelity objects + up to 4 character-consistency images) | May 2026 | |
| Uni 1.1 | Luma AI | $0.0404 (2K text to image), $0.0434 (image edit) — Luma API | 2,048 pixels (2K) | Yes — image edit, with up to 8 image references | May 2026 |
| FLUX.2 Max | Black Forest Labs | From $0.07 (text-to-image and editing), scaling with megapixels | — | Yes — up to 8 reference images (API), 10 (Playground) | May 2026 |
| Recraft V4.1 Pro | Recraft | $0.21 raster, $0.30 vector (SVG) | — | Yes — raster and vector | May 2026 |
| Krea 2 | Krea | $0.03 (Medium); Turbo $0.015, Large $0.06 | 1.5K native (Large 2K); 4K via Enhancer | — | May 2026 |
| GPT Image 2 | OpenAI | Text-to-image (Image API): $0.006/$0.053/$0.211 (low/medium/high quality) at 1024x1024; $0.005/$0.041/$0.165 at 1024x1536 or 1536x1024; additional/arbitrary sizes are billed via image+text token usage rather than a flat per-image price | Up to 3840x2160px (4K landscape) / 2160x3840px (4K portrait) - max edge <=3840px, both edges multiples of 16px, total pixels between 655,360 and 8,294,400; outputs above 2560x1440 (2K) are "experimental"; standard sizes 1024x1024, 1536x1024, 1024x1536 also supported | Yes - up to 16 reference images per request (images array parameter) for editing/reference-based generation; gpt-image-2 always processes image inputs at high fidelity (the input_fidelity parameter is not adjustable for this model, unlike earlier GPT Image models) | Apr 2026 |
| MAI Image 2 | Microsoft AI | $5 text input, $33 image output per 1M tokens | — | — | Apr 2026 |
| Wan 2.7 Image Pro | Alibaba | $0.075 (International) | 4096×4096 for text-to-image; 2048×2048 otherwise | — | Apr 2026 |
| Qwen Image 2.0 Pro | Alibaba | $0.075 (International) | 2048×2048 (512×512 to 2048×2048) | — | Mar 2026 |
| Recraft V4 | Recraft | $0.04 raster, $0.08 vector (SVG) | — | Yes — raster and vector | Mar 2026 |
| Seedream 5.0 Lite | ByteDance | $0.035 per output image; input images free | 4K (2K, 3K or 4K; up to 4096×4096 at 1:1) | Yes — single or multiple reference images | Jan 2026 |
| GLM-Image | Zhipu AI | $0.015 | — | Yes — editing, style transfer, identity-preserving and multi-subject generation | Jan 2026 |
| Grok Imagine Image | SpaceXSI | $0.02 | — | Yes — edits from up to 5 source images | Jan 2026 |
| GPT Image 1.5 | OpenAI | Low $0.009, medium $0.034, high $0.133 at 1024×1024; larger sizes cost more | 1536×1024 or 1024×1536 | Yes — image input and edits endpoint | Dec 2025 |
| Wan 2.6 T2I | Alibaba | $0.03 (International) | 1440×1440 (1280×1280 to 1440×1440) | — | Dec 2025 |
| Seedream 4.5 | ByteDance | $0.04 per output image; input images free | 4K (2K or 4K; up to 4096×4096 at 1:1) | Yes — single or multiple reference images | Nov 2025 |
| FLUX.2 Flex | Black Forest Labs | From $0.05 (text-to-image and editing), scaling with megapixels | — | Yes — up to 8 reference images (API), 10 (Playground) | Nov 2025 |
| FLUX.2 Pro | Black Forest Labs | Text-to-image: $0.03 for the 1st output MP + $0.015/MP thereafter; image editing: from $0.045 (output MP + $0.015/MP per reference image); reference images $0.015/MP each (1 MP = 1024x1024px, rounded up per image) | Up to 4MP output (e.g. 2048x2048 at 1:1); width/height each must be >=64px (API); images/references exceeding 4MP are resized down to 4MP | Yes - text-and-image-to-image editing via input_image through input_image_8 (up to 8 reference images via the API, 10 in the Playground) | Nov 2025 |
| Recraft V3 | Recraft | $0.04 raster, $0.08 vector (SVG) | — | Yes — raster and vector | Jun 2025 |
| FLUX Kontext Max | Black Forest Labs | $0.08 | — | Yes — in-context editing from a reference image | May 2025 |
| GPT Image 1 | OpenAI | Low $0.011, medium $0.042, high $0.167 at 1024×1024; larger sizes cost more | 1536×1024 or 1024×1536 | Yes — image input and edits endpoint | Apr 2025 |
| Ideogram V3 | Ideogram | $0.03 (Flash, Turbo), $0.06 (Default), $0.09 (Quality) | — | Yes — Remix | Mar 2025 |
| Stable Diffusion 3.5 Large | Stability AI | 6.5 credits ($0.065) — Stability AI API | — | — | Oct 2024 |
| Ideogram 4.5 | Ideogram | $0.03 (Low), $0.06 (Medium), $0.10 (High); edits from $0.008 | Native 2K; edits images at their full resolution | Yes — generation with source images | — |
Frequently asked questions
How are image generation APIs priced?+
Most bill a fixed price per image, often higher for larger images or higher quality modes. Some, such as Microsoft’s MAI-Image models, bill per million input and output tokens instead, and a few bill in credits. The table gives each vendor’s own unit.
Is there a free image generation API?+
These image APIs list a free tier or free credits: Gemini Nano Banana 2.1 (Not on the Gemini API free tier); Midjourney V8.2 (Not offered on web or Discord (limited trial via the niji·journey mobile app only)); Qwen Image 3.0 Pro (10 images, valid 90 days); Gemini 3.1 Flash Image (Not available); Gemini 3.1 Flash Image (Nano Banana 2) (Not offered via the Gemini API (free tier explicitly excludes image input/output pricing rows); available at no cost, subject to shared compute-based usage limits, in the consumer Gemini app on every tier including "Without a Google AI plan"); GPT Image 2 (Not offered via the API - the Free usage tier lists image generation as "Not supported" in the model's own rate-limit table, and no free/trial image credits are documented; available at no cost, subject to "limited and slower image generation" rate limits, on ChatGPT's Free plan (the app's current image-generation engine)); Wan 2.7 Image Pro (50 images, valid 90 days); Qwen Image 2.0 Pro (100 images, valid 90 days); Wan 2.6 T2I (50 images, valid 90 days); FLUX.2 Pro (Not offered - no free API or Playground credits found; BFL's own Quick Start guide has new accounts "Add credits to your account...Start with $10-20 to experiment" before generating anything); Stable Diffusion 3.5 Large (25 free API credits for new accounts).
Which companies offer image generation APIs?+
The models on this page come from Black Forest Labs, Google, Microsoft AI, Midjourney, Meta, Alibaba, ByteDance, Ideogram, SpaceXSI, Luma AI, Recraft, Krea, OpenAI, Zhipu AI, Stability AI.
Prices come from each model’s page on this site, which links the vendor’s price list. Video models: AI video generation API pricing. Every model by developer: AI companies.