mirror of
https://github.com/Narcooo/inkos.git
synced 2026-08-28 23:02:03 +08:00
docs: replace keaiapi references with kkaiapi
This commit is contained in:
+4
-4
@@ -28,7 +28,7 @@
|
||||
|
||||
InkOS is an AI Agent system for story creation and multilingual translation: long-form novels, standalone short fiction, scripts, storyboards, fan fiction, spinoffs, style imitation, continuation, interactive film projects, interactive worlds, and long-document translation all start from the same workbench. Studio Chat, CLI, and TUI share the same action surface for discussion, confirmed actions, generation, review, persistent editing, and cross-language delivery.
|
||||
|
||||
> 💡 **One key for global frontier models** — pair InkOS with [**keaiapi**](https://keaiapi.com/): an OpenAI-compatible gateway for Claude, GPT, Gemini, DeepSeek, Kimi, Qwen, GLM, and image models. Add it as a custom service with base URL `https://api.keaiapi.com/v1`, then switch models in Studio without juggling multiple provider accounts.
|
||||
> 💡 **One key for global frontier models** — pair InkOS with [**kkaiapi**](https://en.kkaiapi.com/): an OpenAI-compatible gateway for Claude, GPT, Gemini, DeepSeek, Kimi, Qwen, GLM, and image models. Add it as a custom service with base URL `https://api.kkaiapi.com/v1`, then switch models in Studio without juggling multiple provider accounts.
|
||||
|
||||
## v1.7.0 Multilingual Translation, Cross-Language Creation, and Long-Task Reliability
|
||||
|
||||
@@ -78,7 +78,7 @@ This release continues the v1.5 direction: heavy actions are confirmable, comple
|
||||
|
||||
**Runtime skills and research** — add reusable professional skills under `.inkos/skills/`, force them with `@skill-id`, or ask for web research to generate a sourced Markdown report.
|
||||
|
||||
**Model setup** — Studio includes provider settings, model routing, cover-service settings, [keaiapi](https://keaiapi.com/) / OpenRouter aggregator entries, and custom OpenAI-compatible endpoints.
|
||||
**Model setup** — Studio includes provider settings, model routing, cover-service settings, [kkaiapi](https://en.kkaiapi.com/) / OpenRouter aggregator entries, and custom OpenAI-compatible endpoints.
|
||||
|
||||
<p align="center">
|
||||
<img src="assets/play-item-warcraft.png" width="420" alt="InkOS Play item image example">
|
||||
@@ -161,7 +161,7 @@ inkos
|
||||
|
||||
Open Studio, then go to **Model Settings**:
|
||||
|
||||
1. Choose a service such as Google Gemini, Moonshot, MiniMax, DeepSeek, keaiapi, OpenRouter, or a custom endpoint.
|
||||
1. Choose a service such as Google Gemini, Moonshot, MiniMax, DeepSeek, kkaiapi, OpenRouter, or a custom endpoint.
|
||||
2. Paste the API key and test the connection.
|
||||
3. Pick an available model and save.
|
||||
4. Return to Studio Chat or your book page.
|
||||
@@ -239,7 +239,7 @@ If a service test fails, first check that the service, model, and protocol match
|
||||
### LLM Configuration Notes
|
||||
|
||||
- **Studio / CLI config isolation**: Studio always uses the service page settings and `.inkos/secrets.json`; the CLI, daemon, and deployment environments support env overrides and one-off command flags.
|
||||
- **Provider bank capability table**: built-in baseUrl, protocol, models, and compatibility policies for 15 services — Google Gemini, Moonshot, MiniMax, Zhipu (GLM), Bailian (Alibaba Cloud Model Studio), DeepSeek, SiliconFlow, Volcengine, Tencent Hunyuan, Baidu ERNIE (Wenxin), iFlytek Spark, OpenRouter, keaiapi, Ollama, and CodingPlan.
|
||||
- **Provider bank capability table**: built-in baseUrl, protocol, models, and compatibility policies for 15 services — Google Gemini, Moonshot, MiniMax, Zhipu (GLM), Bailian (Alibaba Cloud Model Studio), DeepSeek, SiliconFlow, Volcengine, Tencent Hunyuan, Baidu ERNIE (Wenxin), iFlytek Spark, OpenRouter, kkaiapi, Ollama, and CodingPlan.
|
||||
- **Model ownership validation**: mismatches like `--service google --model kimi-k2.5` fail immediately, so requests are never sent to the wrong provider.
|
||||
- **Google Gemini compatibility fix**: AI Studio API keys work directly with the Gemini OpenAI-compatible endpoint; InkOS automatically disables the OpenAI `store` parameter Google does not support.
|
||||
- **MiniMax transport probing**: MiniMax / MiniMax CodingPlan use the official OpenAI-compatible `/v1` entry and automatically pick a working non-streaming transport, working around streams that report usage but return an empty body.
|
||||
|
||||
+3
-3
@@ -28,7 +28,7 @@
|
||||
|
||||
InkOS は、物語創作と多言語翻訳のための AI Agent システムです。長編小説、独立短編、脚本、絵コンテ、二次創作、番外、文体模倣、続き書き、インタラクティブ影遊、インタラクティブ世界、長文翻訳を同じワークベンチから始められます。Studio Chat、CLI、TUI は同じ action surface を共有し、相談、確認、生成、レビュー、永続編集、言語をまたぐ納品を一つの流れで扱えます。
|
||||
|
||||
> 💡 **主要モデルをキー 1 本で** —— InkOS には [**keaiapi**](https://keaiapi.com/) の併用がおすすめです。Claude / GPT / Gemini / DeepSeek / Kimi / Qwen / GLM と画像モデルを扱える OpenAI 互換ゲートウェイとして、base URL `https://api.keaiapi.com/v1` をカスタムサービスに設定すれば、複数プロバイダーのアカウントを行き来せずに Studio でモデルを切り替えられます。
|
||||
> 💡 **主要モデルをキー 1 本で** —— InkOS には [**kkaiapi**](https://en.kkaiapi.com/) の併用がおすすめです。Claude / GPT / Gemini / DeepSeek / Kimi / Qwen / GLM と画像モデルを扱える OpenAI 互換ゲートウェイとして、base URL `https://api.kkaiapi.com/v1` をカスタムサービスに設定すれば、複数プロバイダーのアカウントを行き来せずに Studio でモデルを切り替えられます。
|
||||
|
||||
## v1.7.0 多言語翻訳、クロスランゲージ創作、長時間タスクの信頼性
|
||||
|
||||
@@ -78,7 +78,7 @@ v1.6.0 は、InkOS を開放世界 Play からさらにインタラクティブ
|
||||
|
||||
**Runtime Skill とリサーチ** — `.inkos/skills/` に専門 skill を追加し、`@skill-id` で強制使用できます。外部事実が必要な場合は出典付き Markdown リサーチレポートを生成できます。
|
||||
|
||||
**モデル設定** — Studio はサービス設定、モデルルーティング、表紙サービス、[keaiapi](https://keaiapi.com/) / OpenRouter などのモデル集約入口、カスタム OpenAI-compatible エンドポイントに対応します。
|
||||
**モデル設定** — Studio はサービス設定、モデルルーティング、表紙サービス、[kkaiapi](https://en.kkaiapi.com/) / OpenRouter などのモデル集約入口、カスタム OpenAI-compatible エンドポイントに対応します。
|
||||
|
||||
<p align="center">
|
||||
<img src="assets/play-item-warcraft.png" width="420" alt="InkOS Play アイテム画像例">
|
||||
@@ -126,7 +126,7 @@ inkos
|
||||
|
||||
Studio を開き、**モデル設定**へ進みます:
|
||||
|
||||
1. Google Gemini、Moonshot、MiniMax、DeepSeek、keaiapi、OpenRouter、またはカスタムエンドポイントを選択。
|
||||
1. Google Gemini、Moonshot、MiniMax、DeepSeek、kkaiapi、OpenRouter、またはカスタムエンドポイントを選択。
|
||||
2. API Key を貼り付けて接続をテスト。
|
||||
3. 利用可能なモデルを選んで保存。
|
||||
4. Studio Chat または書籍ページに戻って創作を開始。
|
||||
|
||||
Reference in New Issue
Block a user