Example media
Reviewed visual examples
How to use in PerchanceAI
Workflow steps
- 1
Verify the official status
Check the OpenAI launch post and Help Center for access, model IDs, pricing, and preview limitations before writing product-facing copy.
- 2
Map the tier to the job
Use Sol, Terra, and Luna as planning labels for hard reasoning, balanced work, and fast draft workflows rather than fixed benchmark promises.
- 3
Keep PerchanceAI routing honest
Route users to available PerchanceAI image workflows and keep GPT-5.6 support framed as research until catalog evidence confirms it.
Specs
Specs, availability and credits
| Primary sources | OpenAI launch post, Help Center preview article, and Deployment Safety Hub system card Use official sources for IDs, access, pricing, caching, safety evidence, and availability. |
|---|---|
| Model IDs | gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna |
| Preview access | Limited API and Codex preview for approved organizations OpenAI says ChatGPT is not included during preview. |
| Pricing | $5/$30, $2.50/$15, and $1/$6 per 1M input/output tokens Pricing can change, so verify official docs before operational use. |
| PerchanceAI status | Overview only No runtime support claim until Product catalog evidence exists. |
Common mistakes
Avoid these before publishing
- Treating a limited preview as broad ChatGPT or public API availability.
- Presenting Sol, Terra, and Luna as permanent quality rankings instead of preview tiers.
- Claiming PerchanceAI runtime support before Product catalog routing and billing are confirmed.
Next action
Turn model research into a reviewable brief
Use the current PerchanceAI image workflow for brief writing and prompt planning while GPT-5.6 access remains a separate preview topic.
Official preview status
OpenAI describes GPT-5.6 as a limited preview model family, not a broadly available ChatGPT feature during preview. The useful PerchanceAI page answers naming, access, pricing, and planning questions without turning preview access into a product promise.
Model family frame
The three public preview names map to different workflow needs: Sol for the highest-capability frontier tasks, Terra for balanced cost and performance, and Luna for fast high-volume work. Keep these as source-backed descriptions, not permanent rankings.
- Sol: flagship preview tier.
- Terra: balanced lower-cost tier.
- Luna: fastest and lowest-cost tier.
PerchanceAI boundary
Use this page to help creators translate model-release research into practical briefs, prompt tests, and review checklists for currently available PerchanceAI workflows. Do not state that GPT-5.6 is selectable until the live Product catalog proves it.
Prompt examples
Copy-ready starts
Copy an example, then replace the subject and production details with material you can use.
Plan a frontier model evaluation for a product image workflow: define the task, required tools, safety boundary, review criteria, and what evidence would confirm readiness.
Create a balanced AI workflow brief for campaign images with cost controls, prompt structure, review steps, and fallback rules for unsupported model access.
Write a fast visual concept prompt for three ecommerce draft images, clear subject, neutral background, realistic material, and a short review checklist.
FAQ
Frequently asked questions
No. This is an overview page for model research and workflow planning. PerchanceAI should only claim GPT-5.6 support after the Product catalog, routing, pricing, and provider access are confirmed.
OpenAI's preview help article lists `gpt-5.6-sol`, `gpt-5.6-terra`, and `gpt-5.6-luna` for the Sol, Terra, and Luna tiers.
No. OpenAI describes the preview as limited to trusted partners and organizations through approved API organizations and Codex workspaces, with no public waitlist.