Simple, Transparent Pricing

Every request is billed dynamically, with the metric chosen to match each task: resolution × steps for images, characters for speech, tokens for embeddings, duration with optional resolution for video, hours for transcription, input resolution for OCR, and per-image rates for background removal and upscaling. There are no subscriptions, no monthly minimums, and no hidden fees — you fund a prepaid balance and each successful inference deducts its exact cost. Before any job, you can call the matching /price endpoint to preview the precise cost for the model and parameters you plan to use.

Yes. Every new account receives $5 in free credits the moment you sign up — no credit card, no upfront commitment. That balance is enough to test most models extensively, generate hundreds of images, transcribe several hours of video, or prototype a complete AI pipeline. Your account starts on the Basic tier with conservative rate limits designed for testing; making any payment upgrades you to Premium with no waiting period.

Inference is routed through a globally distributed GPU network rather than concentrated in a few hyperscale data centers, which removes most of the infrastructure markup baked into traditional cloud pricing. We also serve highly optimized open-source models — many of them quantized (INT8, FP8, NF4) and distilled — so each request uses fewer GPU seconds without sacrificing output quality. The combined effect can deliver up to 20× lower inference cost for comparable workloads.

deAPI uses Stripe for secure card payments and supported local methods. You can either make one-off top-ups (with preset amounts of $10, $25, or $50) or enable automatic top-ups, which recharge your balance whenever it drops below $2 — perfect for production workloads that can't afford to fail mid-job. For B2B customers needing custom invoices, larger commitments, or tailored billing terms, our team arranges individual agreements; just reach out to support.

The Basic tier offers conservative rate limits (typically 1–10 RPM depending on the endpoint) ideal for testing. The moment you make any payment via Stripe, your account upgrades to Premium with 300 RPM across all endpoints and unlimited daily requests — instantly, with no application process. For high-volume production needs beyond Premium, dedicated capacity, or enterprise terms, our team can set up bespoke arrangements with volume discounts.

Pricing is driven by the number of inference steps required to transform your input image. Starting prices are $0.00659 for FLUX.2 Klein 4B BF16 at 4 steps and $0.03485 for Qwen Image Edit Plus at 40 steps — with both models scaling proportionally as you raise step count. The main lever you control is therefore step count: fewer steps for fast style transfers, more steps for high-fidelity edits. Use the calculator on this page or the /price endpoint to preview exact cost before each request.

Pick Qwen Image Edit Plus when your edit is best described in natural language ("turn the wall terracotta, add a green plant in the corner, switch lighting to golden hour") — it excels at instruction-driven, multi-step compositional edits and defaults to 40 inference steps. Pick FLUX.2 Klein 4B BF16 when you need fast cinematic color grading, atmospheric tweaks, or multi-reference style blending in a sub-second pass at 4 steps.

You can upload JPG, JPEG, PNG, GIF, BMP, or WebP files up to 10 MB per image. Both single-image and multi-image inputs are supported via the image and images[] fields respectively (mutually exclusive). Maximum number of input images depends on the chosen model and is exposed via info.features.max_input_images on the Model Selection endpoint, so always read those limits before sending requests with multiple references.

Light edits like color grading or texture enhancement look great at very low step counts. Moderate transformations — changing scene mood, swapping backgrounds, restyling clothing — typically benefit from mid-range step counts. For complete creative reinterpretations or hero shots in marketing, push toward the higher end of the model's supported range. Always test at lower step counts first; the cost-to-quality curve flattens noticeably above the model's sweet spot, so additional steps mainly add cost.

Yes — FLUX.2 Klein 4B BF16 is a unified model that supports multi-reference editing, letting you blend the style from one image with the composition or subject from another in a single API call. Pass the input images via the images[] field (mutually exclusive with the single image field), and check the model's max_input_images value via the Model Selection endpoint to confirm how many references it accepts.