Amanai Docs
Billing & pricing
One place for billing behavior, current model rates, settlement, and credit packs.
Overview
Amanai uses prepaid credits across chat, media, realtime, embeddings, and rerank. Each endpoint bills in a unit that matches the work delivered; specialized work is not converted into fake chat tokens.
System One structured evaluation costs 0 credits while the offer is active. Currently free for a limited time while the upstream free offering remains available. Free access is not guaranteed and future requests may be priced differently. Usage is still measured.
Current quote wins. Rates can change for future usage as capacity changes. Already settled requests are never recalculated. Use the dashboard calculator or GET /calculate for the quote that applies to a request.
1,000,000,000 credits = Rp150,000 = US$10.
Billing units
| API family | Billing unit | Settlement basis |
|---|---|---|
| Chat | Input, cached input, and output tokens | Usage reported by the completed or streamed response |
| Embeddings | BGE, Qwen3, and Voyage: tokens; EmbeddingGemma: UTF-8 units | Accepted input with a model-specific minimum |
| Rerank | Query + document tokens | All submitted documents, not only top_n |
| System One | Free request | 0 credits; token usage reported |
| Images | Delivered image | Usable output images |
| Video | Requested second | Completed durable job |
| Speech synthesis | Submitted characters | Playable audio delivery |
| Transcription | Verified audio duration | Delivered transcript |
| Realtime | Successfully established session | Successful WebRTC negotiation |
Chat model pricing
Chat credits are rounded up to a whole credit:
credits = ceil(
(input_tokens - cache_read_tokens) × m_in
+ cache_read_tokens × m_cache
+ output_tokens × m_out
) m_cache is the rate for input actually reported as cache-read. Cache creation remains input; only cache reads receive the cache multiplier.
| Model | Input | Output | Cache read |
|---|---|---|---|
amanai/glm-5.3 | 5.6 | 28 | 1.4 |
amanai/glm-5.3-flash | 2.8 | 14 | 0.7 |
amanai/glm-5.2 | 5.6 | 28 | 1.4 |
amanai/glm-5.1 | 5.6 | 28 | 1.4 |
amanai/glm-5.0-turbo | 4.8 | 24 | 1.2 |
amanai/glm-5v-turbo | 4.8 | 24 | 1.2 |
amanai/muse-spark-1.1 | 5 | 25 | 1.25 |
amanai/muse-spark-1.2 | 5 | 25 | 1.25 |
amanai/qwen3.8-max | 6 | 30 | 1.5 |
amanai/qwen3.7-max | 4.425 | 22.125 | 1.10625 |
amanai/qwen3.7-plus | 2.24 | 11.2 | 0.56 |
amanai/kimi-k3 | 9 | 45 | 2.25 |
amanai/kimi-k2.7 | 3.8 | 19 | 0.95 |
amanai/kimi-k2.6 | 3.8 | 19 | 0.95 |
amanai/kimi-k2.5 | 2.28 | 11.4 | 0.57 |
amanai/deepseek-v4-pro | 5 | 25 | 1.25 |
amanai/deepseek-v4-pro-0813 | 5 | 25 | 1.25 |
amanai/deepseek-v4-flash | 2.5 | 12.5 | 0.625 |
amanai/deepseek-v4.1-flash | 2.5 | 12.5 | 0.625 |
amanai/deepseek-v4-flash-0731 | 2.5 | 12.5 | 0.625 |
amanai/minimax-m3 | 2.1 | 10.5 | 0.525 |
amanai/minimax-m2.7 | 1.2 | 6 | 0.3 |
amanai/hy3 | 4.5 | 22.5 | 1.125 |
amanai/hy4 | 4.5 | 22.5 | 1.125 |
amanai/grok-4.5 | 6 | 30 | 1.5 |
amanai/grok-4.6 | 6 | 30 | 1.5 |
amanai/gpt-5.5 | 10 | 50 | 2.5 |
amanai/gpt-6-astra | 30 | 150 | 7.5 |
amanai/gpt-5.6-sol | 10 | 50 | 2.5 |
amanai/gpt-5.6-terra | 7.5 | 37.5 | 1.875 |
amanai/gpt-5.6-luna | 5.625 | 28.125 | 1.40625 |
amanai/claude-fable-5.1 | 30 | 150 | 7.5 |
amanai/claude-fable-5 | 30 | 150 | 7.5 |
amanai/claude-opus-5 | 15 | 75 | 3.75 |
amanai/claude-opus-4.8 | 15 | 75 | 3.75 |
amanai/claude-opus-4.7 | 15 | 75 | 3.75 |
amanai/claude-opus-4.6 | 15 | 75 | 3.75 |
amanai/claude-sonnet-5 | 10.5 | 52.5 | 2.625 |
amanai/claude-sonnet-4.6 | 9 | 45 | 2.25 |
amanai/claude-haiku-4.5 | 3 | 15 | 0.75 |
Embedding model pricing
Token-metered embeddings: credits = max(250, ceil(tokens × model rate)). EmbeddingGemma remains byte-metered: credits = max(1,000, ceil(UTF-8 units × rate)). Local byte counts are not exact token counts.
| Model | Dimensions | Rate | Minimum |
|---|---|---|---|
amanai/bge-small-en | 384 | 1.25 credits / token | 250 credits |
amanai/bge-base-en | 768 | 4 credits / token | 250 credits |
amanai/bge-large-en | 1024 | 12.5 credits / token | 250 credits |
amanai/bge-m3 | 1024 | 0.75 credits / token | 250 credits |
amanai/embeddinggemma-300m | 768 | 1 credit / UTF-8 unit | 1,000 credits |
amanai/qwen3-embedding-0.6b | 1024 | 0.75 credits / token | 250 credits |
amanai/voyage-4-lite | 1024 | 1.25 credits / token | 250 credits |
amanai/voyage-4 | 1024 | 3.75 credits / token | 250 credits |
amanai/voyage-4-large | 1024 | 7.5 credits / token | 250 credits |
amanai/voyage-code-4 | 1024 | 7.5 credits / token | 250 credits |
amanai/voyage-finance-2 | 1024 | 7.5 credits / token | 250 credits |
amanai/voyage-law-2 | 1024 | 7.5 credits / token | 250 credits |
Rerank pricing
BGE and Voyage count tokens for the query and every submitted document. The rates and minimums below are model-specific; top_n never reduces billed input.
| Model | Rate | Minimum | Counted input |
|---|---|---|---|
amanai/bge-reranker-base | 0.25 credits / token | 250 credits | Query plus every submitted document; top_n does not reduce billed input |
amanai/voyage-rerank-3 | 3 credits / token | 250 credits | Query plus every submitted document; top_n does not reduce billed input |
amanai/voyage-rerank-3-lite | 1.25 credits / token | 250 credits | Query plus every submitted document; top_n does not reduce billed input |
amanai/voyage-rerank-2.5 | 3 credits / token | 250 credits | Query plus every submitted document; top_n does not reduce billed input |
amanai/voyage-rerank-2.5-lite | 1.25 credits / token | 250 credits | Query plus every submitted document; top_n does not reduce billed input |
System One pricing
Jev returns structured decisions at 0 credits per request while the offer is active. Currently free for a limited time while the upstream free offering remains available. Free access is not guaranteed and future requests may be priced differently. Input and output token usage is still included in the response and request logs.
| Model | Vendor | Rate | Billing |
|---|---|---|---|
amanai/jev-1.13 | TypeSafe AI | Free · 0 credits | Per request while offer is active |
Media model pricing
| Model | Capability | Current rate | Settlement |
|---|---|---|---|
amanai/gpt-5.6-luna-image | Image generation | 5,000,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/gpt-5.5-image | Image generation | 7,500,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/gpt-5.6-terra-image | Image generation | 7,500,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/gpt-5.6-sol-image | Image generation | 10,000,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/grok-imagine-image | Image generation and editing | 3,000,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/grok-imagine-image-2.0 | Image generation and editing | 4,000,000 credits / delivered image | Delivered images only; unusable output is refunded |
amanai/grok-imagine-video | Video generation | 7,000,000 credits / second at 720p | Reserved at submission; settled or refunded once |
amanai/grok-imagine-video-1.5 | Video generation | 10,500,000 credits / second at 720p | Reserved at submission; settled or refunded once |
amanai/grok-voice-tts | Speech synthesis | 2,000,000 credits / 1,000 characters · 500,000 minimum | Settled after playable audio is delivered |
amanai/grok-voice-transcribe-1.0 | Speech transcription | 400,000 credits / minute · 250,000 minimum | Settled after a transcript is delivered |
amanai/grok-voice-transcribe-2.0 | Speech transcription | 600,000 credits / minute · 250,000 minimum | Settled after a transcript is delivered |
Image output modifiers
| Selection | Multiplier | Applies to |
|---|---|---|
quality=low | 0.75× | GPT image models |
quality=auto or medium | 1× | GPT image models |
quality=high | 1.5× | GPT image models |
1536x1024 or 1024x1536 | 1.5× | GPT image models |
| Each additional edit input image | +0.25× | Where multi-image editing is supported |
| Grok Image 2.0 · low / 1K | 1× | Generation default |
| Grok Image 2.0 · low / 2K or medium / 1K | 1.5× | Output image |
| Grok Image 2.0 · medium / 2K | 2× | Output image |
| Grok Image 2.0 edit source | +0.25× | Per accepted source image |
For Grok Image 2.0, quality=auto is priced as low for generation and medium for editing. The modifier is based on the selected quality and resolution; aspect_ratio does not add a separate charge. The model's base price and the edit-source surcharge still apply.
Realtime pricing
| Model | Current rate | Settlement |
|---|---|---|
amanai/gpt-live | 7,000,000 credits / successful session | Failed session negotiation is not billed |
Reserve and settlement
Before a costed call begins, the gateway atomically reserves a bounded worst-case amount. After delivery it settles the actual charge and returns any unused reserve.
Credit packs
Credits are prepaid and valid for 30 days. Account limits are shared across keys unless a key has a tighter scope.
| Pack | Credits | QRIS | USDC | RPM | Concurrency |
|---|---|---|---|---|---|
| Pro | 250,000,000 | Rp60,000 | 4 USDC | 120 | 8 |
| Max | 500,000,000 | Rp90,000 | 6 USDC | 200 | 16 |
| Ultra | 1,000,000,000 | Rp150,000 | 10 USDC | 320 | 24 |
Check the active balance with:
curl https://api.amanai.dev/v1/usage -H "Authorization: Bearer $AMANAI_API_KEY"