ヘルプとwiki お絵かき サインイン

Japanese Storywriting AI
AI Novelist 2.0
by Bit192

Developer API


Integrate AI Novelist in your app.
* Currently in limited alpha. We are not supporting large-scale public deployment (except personal client/app) yet.


API Key





How to Use


Read [ API Docs ] for REST API and quickstart!

If you need a support, check the corresponding thread in #ai-forums in [ Bit192 Discord ]




Pricing


Context Generation
spiko_ultra
Spiko Ultra		 0.275 Lumina (≒0.33JPY)
/ 1000 tokens		 2.25 Lumina (≒2.79JPY)
/ 1000 tokens
derrida_03
Derrida-03 (thinking)		 0.44 Lumina (≒0.54JPY)
/ 1000 tokens		 1.75 Lumina (≒2.17JPY)
/ 1000 tokens
spiko_max
Spiko-sama Max (thinking)		 0.44 Lumina (≒0.54JPY)
/ 1000 tokens		 2.5 Lumina (≒3.1JPY)
/ 1000 tokens
spiko
Spiko-sama-02		 0.44 Lumina (≒0.54JPY)
/ 1000 tokens		 1.75 Lumina (≒2.17JPY)
/ 1000 tokens
spiko_solid
Spiko-sama-02 / Solid		 0.44 Lumina (≒0.54JPY)
/ 1000 tokens		 1.75 Lumina (≒2.17JPY)
/ 1000 tokens
damsel_ray
Damsel+RAY		 0.67 Lumina (≒0.83JPY)
/ 1000 tokens		 2.5 Lumina (≒3.1JPY)
/ 1000 tokens
Legacy Models
supertrin_highpres
SuperTrin Extended / High Precision		 0.4 Lumina (≒0.496JPY)
/ 1000 tokens		 1.75 Lumina (≒2.17JPY)
/ 1000 tokens
supertrin_maxpres
SuperTrin Extended / Max Precision		 0.4 Lumina (≒0.496JPY)
/ 1000 tokens		 2.5 Lumina (≒3.1JPY)
/ 1000 tokens
supertrin
SuperTrin Extended		 0.4 Lumina (≒0.496JPY)
/ 1000 tokens		 1.5 Lumina (≒1.86JPY)
/ 1000 tokens
damsel
Damsel 20B		 0.7 Lumina (≒0.86JPY)
/ 1000 tokens		 1.75 Lumina (≒2.17JPY)
/ 1000 tokens

Natively Supported Languages OpenAI-Compatible
Endpoint		 Max. Input Max. Output Rate Limit
spiko_ultra
Spiko Ultra		 Japanese, English, Simplified Chinese, Traditional Chinese, Korean ◯
v1/completions
v1/chat/completions		 200,000 32,768 300 Requests / Minute
derrida_03
Derrida-03 (thinking)		 Japanese, English, Simplified Chinese, Traditional Chinese, Korean × 40,000 8,192 300 Requests / Minute
spiko_max
Spiko-sama Max (thinking)		 Japanese, English, Simplified Chinese, Traditional Chinese, Korean × 40,000 81,92 300 Requests / Minute
spiko
Spiko-sama-02		 Japanese, English, Simplified Chinese, Traditional Chinese, Korean × 40,000 8,192 300 Requests / Minute
spiko_solid
Spiko-sama-02 / Solid		 Japanese, English, Simplified Chinese, Traditional Chinese, Korean × 40,000 8,192 300 Requests / Minute
damsel_ray
Damsel+RAY		 Japanese × 12,288 400 300 Requests / Minute
Legacy Models
supertrin_highpres
SuperTrin Extended / High Precision		 Japanese × 9,216 400 300 Requests / Minute
supertrin_maxpres
SuperTrin Extended / Max Precision		 Japanese × 9,216 400 300 Requests / Minute
supertrin
SuperTrin Extended		 Japanese × 9,216 400 300 Requests / Minute
damsel
Damsel 20B		 Japanese × 2,400 400 90 Requests / Minute

* Purchase extra Lumina [ here ] .


Prompt Caching


The API server automatically caches frequently submitted prompt for up to 10 minutes to speed up the processing. All models are prompt caching enabled by default (except legacy models). This is baked in to our API input price based on average cache hit rate.


Ext. Libraries


The libraries below are compatible with AI Novelist's developer API. Please note that these libraries are community projects that are supported and maintained by each individual developers.

GUI Client

JavaScript

VS Code

WOLF RPGエディター



Terms of Service


Please refer to our [ Terms of Use ] .


デリダ