IDR first
Usage is presented directly in Indonesian Rupiah so product teams can forecast in the currency their business operates in.
12 / PRICINGRAYA.COM.AI
Token pricing designed for commercial use.
Preview pricing is denominated in Indonesian Rupiah, with separate rates for input and output Tokens across each localized model profile, plus dedicated enterprise options.
POSITION
A local model platform should be commercially legible in its home market. RAYA publishes an indicative IDR price architecture now, while clearly separating preview rates from final general-availability terms.
PREVIEW RATE CARD / IDRPER 1M TOKENS · SUBJECT TO CHANGE
MODEL PROFILEINPUTOUTPUTACCESS
RAYA Local / CompactRp3,000Rp6,000Shared API
RAYA Local / GeneralRp8,000Rp24,000Shared API
RAYA Local / ReasonRp10,000Rp40,000Shared API
RAYA Embed / MultilingualRp1,500—Embedding API
INDICATIVE PRICING
Rounded preview rates for product planning. Model availability and final commercial terms will be confirmed before general availability.
LOCAL COMMERCIAL DESIGN
Priced for Indonesia
Market calibrated
Preview rates were informed by published API list prices from established model providers and Bank Indonesia reference exchange rates.
Separate meters
Input, output, cached context where supported, and embeddings are measured separately to keep usage legible.
Enterprise terms
Committed usage, dedicated capacity, customization, and private delivery are scoped around the actual workload.
REFERENCE SNAPSHOT / 31 JUL 2026
Rounded preview rates are a proposed RAYA price architecture—not a direct resale or pass-through of any single provider. Final availability, taxes, limits, and commercial terms will be confirmed before general availability.
01 / APPROACH
How it works
Choose a model profile
Different localized profiles may have different quality, reasoning, throughput, and serving-cost characteristics.
Measure model usage
Standard API usage is measured through input, cached input where supported, and generated output tokens.
Apply commercial terms
Published rates, volume arrangements, batch processing, or committed usage can support different adoption patterns.
Configure dedicated delivery
Dedicated capacity, private model profiles, and customization are scoped around the required operating model.
02 / SCOPE
What it includes
Input tokens
Text, instructions, and context sent to a generation model; priced per one million Tokens
→Output tokens
Text generated by the model, including reasoning output where applicable; metered separately from input
→Cached input
Reusable prompt context may receive a lower rate where the selected profile supports caching
→Embeddings
Indonesian and multilingual text converted into vector representations; priced by processed Tokens
→Dedicated capacity
Reserved serving arrangement based on model profile, required capacity, service level, and commitment
→Private model
Customization, deployment boundary, operations, and support quoted for the enterprise requirement
→03 / QUESTIONS
Frequently asked
Are these final production prices?
No. They are indicative preview rates for product planning. Final model availability, rate cards, limits, taxes, and commercial terms will be confirmed before general availability.
How were the IDR rates calibrated?
RAYA reviewed published API list prices from established model providers and converted reference values using Bank Indonesia exchange rates. The rounded RAYA rates are a proposed local price architecture, not a direct resale or pass-through of any one provider.
How will Token usage be calculated?
Usage separates model input, cached input where supported, generated output, and embedding usage, expressed per one million Tokens.
Will enterprise pricing differ?
Yes. Dedicated capacity, private deployment, customization, support, and committed usage require commercial terms beyond standard shared API rates.