P-Image LoRA · Text to Image
Text to image
Current price
$0.006 / image
Product use cases · In-product generation · Images and video
Create companion experiences that feel visually consistent.
Start with an approved character pack, then compare models that accept reference images using the same test scenes. Review the estimate before every run; where needed, choose low-filter capability only on models that clearly offer it.
- Best for
- Product teams adding fictional character images, short video moments and a recognizable visual identity to companion apps.
- What you get
- A reusable character pack and evaluation process that keeps identity more consistent across images and video.
Concept illustration
A creative brief to try
Build a café-to-home welcome
Plan a fictional 30-year-old companion in a navy satin shirt: a café portrait, a relaxed home scene, and an optional greeting clip. Everyday conversation can stay in text.
- 01
Create a clear café portrait with one cup and soft window light. Keep it as the main character reference.
- 02
Return to that portrait for the home scene. Change the setting to a warm living room while keeping the character and shirt.
- 03
If the app needs a moving welcome, animate a chosen portrait with a smile and small wave. Add images only where they help the conversation.
Before you keep the result
Compare the face, hair, shirt, and hands across both scenes. Watch the whole greeting before using it in the app.
Start with your own reference images, follow the scene plan and review each result.
Open this scene in the Playground →Recommended models
169 matching models
Matches use the tasks and required capabilities published in the catalogue. Availability and current prices come from each model page.
FLUX.1 Schnell LoRA · Image Editing
Image editing
Current price
$0.006 / image
FLUX.1 Schnell LoRA · Text to Image
Text to image
Current price
$0.006 / image
Current price
$0.0093 / image
Z-Image · Text to Image
Text to image
Current price
$0.01 / image
Current price
$0.011 / output second
Current price
$0.011 / image
Current price
$0.011 / output second
What one run costs
Worked out from the live catalogue, at the settings named under each figure.
Catalog estimates refresh hourly. Check the estimate for your settings before running; final charges follow the model’s usage and billing policy.
How to build it
Keep the approved character pack in your own persistent storage and reuse accessible HTTPS references across turns. Refresh temporary uploads when they expire, then deliver each completed image through a callback so the chat stays responsive.
- 01
Ask for an upload ticket
POST
/api/v1/common/upload-urlThe ticket is valid only for the byte count you declared: JPEG, PNG, WebP and GIF up to 10 MiB each. Media that already has a public HTTPS URL skips this step entirely.
- 02
Commit the uploaded file
POST
/api/v1/files/{fileId}/commitThe commit checks the bytes and hands back a spicy:// reference, which is what a model's media fields accept. Committing the same file again changes nothing.
- 03
Create the task
POST
/api/v1/jobs/createTask202 means accepted, not finished. Send an Idempotency-Key, and if the price must not move, send the jobs/quote result together with the amount you agreed to.
- 04
Or let the result come to you
POST
{callBackUrl}Leave a callBackUrl and the finished record is delivered there. Verify the signature before you trust the body, and assume the same delivery will arrive twice.
- 05
Move the file into your own storage
POST
/api/v1/common/download-urlMint a short-lived download URL for the output and copy it into your own bucket within the same job. Results stay on our CDN for 14 days.
Copy a request
This example starts with the scene described on this page. Fields come from the selected endpoint’s schema. Replace any example.com reference URL with your own image or committed file URI before sending.
P-Image LoRA · Text to Image →
Where this usually snags
Platform behaviour that is cheaper to design around than to discover in production.
The result URL is not your archive
Generated media stays on our CDN for 14 days and uploaded sources for one day. Copy whatever you keep into your own bucket inside the same job.
The catalogue number is where the price starts
Most video endpoints bill by resolution and length, so the figure beside a model is its lowest tier, not your settings. Budget against the estimate shown before the run.
Give the key a ceiling before it ships
Every API key carries a daily, a monthly and a lifetime spend cap. Set them when you create the key: a leaked key without one spends the balance, not the budget.
Launch checklist
- 01Create an approved character pack and record the fictional-adult or documented-consent status of every likeness.
- 02Compare reference-capable models with the same inputs and representative scenes.
- 03Review the estimate and limit variants before submitting; an accepted task cannot be canceled.
- 04Move results to controlled storage and tell users how long source files and outputs are kept.
Models that match this workflow
Explore model families for this workflow. Open a model page to review its inputs, pricing and current availability.
Frequently asked questions
169 published models match the listed tasks and required capabilities. Compare their inputs, examples, availability and current prices before choosing one.
Establish a character reference image, then reuse it as a first frame or reference input. This helps reduce the variation that comes from recreating the character from text on every turn. Test consistency with the model you plan to use.
Choose an uncensored model that supports unrestricted creative work. SpicyAPI adds no content classifier, per-request declaration or approval queue. Check the model page for its capabilities and input requirements. Your product remains responsible for user access, consent and applicable terms.
Estimate chat, image and video usage separately, then apply each model’s current rates. Video cost can depend on resolution and other settings as well as duration, so it is not always proportional to clip length. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.
Generated media is served from our own CDN for 14 days; uploads are retained for 1 day. We do not automatically publish customer content, but anyone with a valid result link may be able to access the file. Protect those links and copy files you need to keep into your own storage.
About this workflow
Combine LLM dialogue, character images and video through one account, API key and balance. Reusing reference images helps preserve character consistency. Check each model page for supported inputs, availability and billing units. Failed or timed-out tasks are usually not charged. Some models charge for content refusals as stated on their model page; check those billing terms before submitting.
