Image generation and editing. Official billing uses input and output tokens; it is not a fixed per-image price.
Official providers appear first. These are source-listed endpoints, not a ranking: resolutions, modes and billing units can differ.
USD 5 / million_tokenstext input
USD 1.25 / million_tokenscached text input
USD 8 / million_tokensimage input
USD 2 / million_tokenscached image input
USD 30 / million_tokensimage output
USD 0.08 / per imagegpt-image-2-5-flare, image-to-image, 4K
USD 0.05 / per imagegpt-image-2-5-flare, image-to-image, 2K
USD 0.03 / per imagegpt-image-2-5-flare, image-to-image, 1K
USD 0.08 / per imagegpt-image-2-5-flare, text-to-image, 4K
USD 0.05 / per imagegpt-image-2-5-flare, text-to-image, 2K
USD 0.03 / per imagegpt-image-2-5-flare, text-to-image, 1K
USD 0.000008 / input_image / tokenOpenAI; input and output charges can add together.
USD 0.000005 / input_text / tokenOpenAI; input and output charges can add together.
USD 0.00003 / output_image / tokenOpenAI; input and output charges can add together.
USD $0.25 / output imagemodel variant: auto · or 40 images for $10
USD $0.012 / output imagemodel variant: low · or around 83 images for $1
USD $0.047 / output imagemodel variant: medium · or around 21 images for $1
USD $0.128 / output imagemodel variant: high · or around 78 images for $10
USD $0.25 / output imagemodel variant: xhigh · or 40 images for $10
USD $0.50 / output imagemodel variant: max · or 20 images for $10
USD 0.000005 / inputTokenText input
USD 0.000008 / inputTokenImage input
USD 0.00003 / outputTokenImage output
USD 0.0390 / source-listed starting priceDisplayed catalog starting price only. Final charge depends on duration, resolution, outputs and references. Not a fixed per-second or per-image quote.
USD 0.0240 / source-listed starting priceDisplayed catalog starting price only. Final charge depends on duration, resolution, outputs and references. Not a fixed per-second or per-image quote.
gpt-image-2.5-flareAn actual Vidily image generation, 1536 × 1024, high quality. Editorial artwork, not a video-model benchmark.
Original editorial still life for Vidily, a video API price comparison publication. A sculptural dark aubergine film reel rests beside three folded lemon-yellow paper strips of different lengths, a small stack of lilac tokens, and a coral mechanical counter with no digits. The strips create a confident diagonal through the composition, suggesting video duration and a finite budget. Tangible matte paper, crisp cut edges, subtle ceramic surfaces, realistic restrained studio shadows, sophisticated product photography rather than cartoon icons. Pale lilac seamless background, plum, lemon yellow and coral palette. Asymmetric close composition, subject occupying two thirds with useful breathing room. One cohesive scene, not a collage or UI mockup. Landscape 3:2. No words, logos, numbers, fake screenshots, watermarks or human figures. This is editorial brand artwork, not an example output from a video model.
More provider-selected examples on the original site. These are demonstrations, not a controlled Vidily benchmark.
Source-listed endpoints and pricing conditions. Variants are not unique base models; account quotes do not enter cheapest-price rankings.
Counts describe collected endpoints, not unique models or a guarantee of complete coverage. Price evidence may be a numeric rate or source billing text. Providers with no published endpoints are still awaiting collection.
16 endpoints · 1/1
Variants listed separately. Open a model to check units and conditions.
OpenAI · image
IMAGE APIgpt-image-2.5-flareOfficial image generation and editing. Token usage varies by input, output and quality. No flat per-image estimate.
fal.ai · image
IMAGE APIopenai/gpt-image-2.5/flare/editPrecise image editing that changes only what's asked, keeping subject, composition, and background intact, with reference subjects staying recognizable across styles and successive edits.
fal.ai · imageIMAGE APIopenai/gpt-image-2.5/flare/text-to-imageOpenAI's default image model for most applications. Fast, high-quality generation with natural lighting, rich textures, and support for complex layouts including transparent backgrounds.
Runway · image
IMAGE APIgpt_image_2_5_flare1–76 credits per image, by quality and resolution, plus 1 credit per reference image (see below)
gpt-image-2-5-flare, image-to-image, 4Kgpt-image-2-5-flare, image-to-image, 4K
gpt-image-2-5-flare, image-to-image, 2Kgpt-image-2-5-flare, image-to-image, 2K
gpt-image-2-5-flare, image-to-image, 1Kgpt-image-2-5-flare, image-to-image, 1K
gpt-image-2-5-flare, text-to-image, 4Kgpt-image-2-5-flare, text-to-image, 4K
gpt-image-2-5-flare, text-to-image, 2Kgpt-image-2-5-flare, text-to-image, 2K
gpt-image-2-5-flare, text-to-image, 1Kgpt-image-2-5-flare, text-to-image, 1K
gpt-image-2.5-flaregpt-image-2.5 · tiered_token
openai:gpt-image@2.5-flareGPT-Image-2.5 Flare
Replicate · image
IMAGE APIopenai/gpt-image-2.5-flareOpenAI's fastest model for high-quality, everyday image generation. Generate and edit images from text and image inputs with strong instruction following and sharp text rendering.
OpenRouter · image
IMAGE APIopenai/gpt-image-2.5-flareGPT Image 2.5 Flare is an image generation and editing model from OpenAI, positioned as the speed-oriented tier of the GPT Image 2.5 series. It is suited to high-volume everyday...
openai/gpt-image-2.5-flare/editOpenAI's GPT Image 2.5 Flare Edit edits one or more reference images from natural-language instructions, with five quality tiers up to 4K. Flare is the fast, balanced GPT Image 2.5 tier for everyday g
openai/gpt-image-2.5-flare/text-to-imageOpenAI's GPT Image 2.5 Flare Text-to-Image generates high-quality images from natural-language prompts, with five quality tiers up to 4K. Flare is the fast, balanced GPT Image 2.5 tier for everyday ge
Choose the exact output configuration above. Estimates use public source rates and show the applicable billing unit. The total is not a guaranteed checkout price.
A model listing proves availability, not a rate. Login-only pricing, missing conditions or stale evidence must be checked before comparison.
No. Comparable estimates are sorted by cost. Advertising belongs in separately labeled positions.