AI Services in API & SDK

Services that expose functionality as APIs and SDKs for developersFeaturing 28 AI services in total.

キミ

Kimi

Kimi is an AI chat and agent service from Beijing-based Moonshot AI. Its flagship model Kimi K3.1 Pro (announced October 24, 2026; 1M-token context; weights due by November 4) handles slide creation, Deep Research, coding, and 3D creation. There is a free plan, paid plans start with Plus at $19/month, and the K3.1 Pro API costs $2 input / $10 output per 1M tokens. The web version has a Japanese display.

# 生成AI（LLM）# AIチャットボット# テキスト生成# コード生成# AIエージェント# API・SDK

シーダンス

Seedance

Seedance is an AI video generation model developed by ByteDance (China). The latest Seedance 2.5 supports up to 30 seconds of continuous generation and up to 50 reference assets (images, video, audio). General users access it via Dreamina international or CapCut; developers use the BytePlus API. Free daily credits are available, with USD-based credit plans and pay-as-you-go API pricing.

# 動画生成# 品質向上・クリエイティブ支援# API・SDK# アイデア支援／創造ツール# マーケティング

ワン

Wan

Wan (Tongyi Wanxiang) is Alibaba's family of AI video and image generation models with an official online service. Wan2.1/2.2 are open models under Apache 2.0, and the latest Wan 3.0 (beta) supports native 30-second generation. Free daily credits on wan.video, with paid plans from $5/month (billed yearly); billing is in US dollars.

# 動画生成# 画像生成# 品質向上・クリエイティブ支援# API・SDK# アイデア支援／創造ツール

フラックス

FLUX

A family of frontier image-generation and editing models from Germany's Black Forest Labs (founded by Stable Diffusion's creators). The current still-image workhorse is the FLUX.2 generation, with the multimodal FLUX 3 as the newest. Strong at photorealism, prompt adherence, in-image text, and editing (Kontext); usable via Playground/API/open weights on pay-as-you-go (1 credit = $0.01).

# 画像生成# デザイン生成# API・SDK

ディファイ

Dify

Dify is an open-source LLM app development platform for building, deploying, and operating AI apps, agents, and workflows with no-code/low-code. You can assemble chatbots, RAG, and multi-step workflows visually, and switch between the major providers' models. It supports both a cloud edition and self-hosting (OSS). Cloud pricing is Sandbox free / Professional $590 per year / Team $1,590 per year / Enterprise custom (17% off with annual billing), and the self-hosted Community edition is free.

# AIチャットボット# AIエージェント# ノーコード自動化# API・SDK# 業務効率化# 日本語対応

ギットハブ コパイロット

GitHub Copilot

GitHub Copilot is an AI coding assistance service provided by GitHub (a Microsoft subsidiary). In addition to code completion, it includes an agent mode that implements autonomously, a cloud agent that works in the cloud, automatic PR review, and the Copilot CLI. It can be used in VS Code, JetBrains, GitHub, the terminal, and more, and you can choose from major models such as Claude, GPT, and Gemini. For individuals there is the free Free plan, Pro $10, Pro+ $39, and Max $100; for businesses there are Business $19 and Enterprise $39 (per user/month, on a GitHub AI credit system).

# コード生成# AIモデル構築・管理# API・SDK# AIエージェント

グーグル エーアイ スタジオ

Google AI Studio

Google AI Studio is Google's developer platform to try the Gemini models in the browser, build and deploy AI apps, and get a Gemini API key. It's organized around Build (app generation), Chat, Stream, and Generate Media (image/video/audio), and offers the Gemini 3 family, Nano Banana image, Veo video, and more. AI Studio itself is free; the API is usage-based with a free tier. Note: on the free tier input is used to improve products; on the paid tier it is not.

# 生成AI（LLM）# AIモデル構築・管理# API・SDK# AIエージェント

クライン

Cline

Cline is an open-source AI coding agent that runs inside the editor and terminal you already use (built by Cline Bot Inc., Apache 2.0). You can use the same agent from a VS Code extension, JetBrains, a CLI, or an SDK, and with Plan/Act it separates planning from execution so you approve as it goes, autonomously handling multi-file edits and command execution. It supports every provider — Claude, GPT, Gemini, local models and more — and with bring your own key there is no vendor lock-in. The extension and CLI are open source and free for individuals (the only cost is usage-based model inference). Used by more than 8 million developers, with 66,000 GitHub stars.

# コード生成# AIモデル構築・管理# API・SDK# AIエージェント

ジェヴ

Jev

Jev is the first System One model from TypeSafe AI, released on 15 September 2026. Unlike an LLM it generates no text: give it a state (text or JSON) and it returns typed values — Choice (classification), Score (rating) and Noul (a 0–1 truth value) — each with probabilities and a calibrated confidence. Because no string comes back, there is nothing to parse and nothing to validate. Vendor-published figures are 70–500 ms end to end and $0.042 per million input tokens, with output tokens free. Questions in one request are evaluated in parallel, so stacking them barely changes latency — which suits support triage, LLM guardrails and RAG re-ranking, where the same judgement repeats at volume. Still in early access behind a waitlist as of October 2026 and also listed on OpenRouter; the docs state that CJK scripts are accepted with lower accuracy than English. On 9 October 2026 Microsoft released Microsoft-Decision-1 with the same price and question types.

# API・SDK# AIエージェント# 検索エンジン・RAG# 自動化・省力化# コスト削減

シードリーム

Seedream

Seedream is an image generation and editing AI model family developed by ByteDance's Seed research team. The latest Seedream 5.0 excels at precise editing including layer separation and multilingual text rendering. Individuals can use it via Dreamina (free daily credits, Japanese UI); developers via the BytePlus API from $0.035 per output image (USD billing).

# 画像生成# デザイン生成# API・SDK# 品質向上・クリエイティブ支援# 日本語対応

キロコード

Kilo Code

Kilo Code is an MIT-licensed open-source AI coding agent that runs the same way in VS Code, JetBrains, the CLI, and the cloud. You switch between purpose-built agents — Code, Plan, Ask, Debug, and Review — to generate code, refactor, chase bugs, and review changes. It supports 500+ models with mid-task switching, and charges no markup on AI inference: you pay the provider's rate. You can bring your own API keys or run local models with Ollama. Platform access is free for individuals and $15 per user per month for Teams. Anaconda acquired the company in July 2026, and Kilo Code now has more than 3 million users.

# コード生成# AIエージェント# API・SDK# 生産性向上# 日本語対応

アジェンタ

Agenta

An open-source (MIT) workspace for building and running AI agents. Build agents by chatting, connect them to your apps, and run them in the background with approval flows, tracing, and version history. Positioned as "the open-source workspace for your agents," it supports 1,000+ integrations (MCP/Composio) and is model-agnostic. Self-hosting is free, with a cloud version available.

# AIモデル構築・管理# AIエージェント# API・SDK# 生成AI（LLM）

スピークイージー

Speakeasy

A developer platform that auto-generates type-safe, production-ready SDKs in multiple languages (TypeScript, Python, Go, Java, C#, PHP, Ruby) from an OpenAPI spec. It also generates MCP servers, Terraform providers, and CLIs, and in 2026 has expanded into an "AI control plane" for discovering, securing, and governing AI agents and MCP servers across the enterprise.

# API・SDK# コード生成# AIエージェント

クレイドエーアイ

Claid.ai

A product-photography platform (Claid.ai by Let's Enhance) that generates and enhances e-commerce product images and fashion photos with AI. It supports AI photoshoots, background generation/removal, on-model fashion shots, image-to-video, up to 4K upscaling, and light/color correction, and can automate bulk processing via a web app and an API.

# 画像生成# デザイン生成# 小売・EC# API・SDK

レストビーエーアイ

Restb.ai

A computer-vision AI company for real estate that analyzes property photos to extract 'visual insights' — auto-tagging of rooms and features, property-condition analysis, photo compliance, watermark/duplicate detection, comparable properties, and image captions. It provides the visual-intelligence layer behind MLSs, portals, AVMs/appraisals, and insurers. Pricing is not public (contact/API).

# 不動産# 顔認証・画像解析# API・SDK

マストラ

Mastra

An open-source AI agent development framework written in TypeScript. It integrates Agents/Tools/Workflows/RAG/Memory/Evals/Observability and lets you switch among LLMs like OpenAI, Anthropic, and Google Gemini. Under "build, observe, and improve," it suits building and running agents that run for days. The framework is free (OSS), with a hosted Mastra Platform available.

# コード生成# AIモデル構築・管理# API・SDK# AIエージェント

トラエ

Trae

An AI-native IDE developed by ByteDance. Built on VS Code, it lets you switch between an IDE mode you control yourself and a SOLO mode where AI autonomously handles everything from requirements to implementation, testing, and deployment. It has the GPT-5 series, Gemini 3 series, Kimi, MiniMax, and ByteDance Seed built in, with Claude and others addable as custom models via API key. Pricing is token-based, from a free tier up to Ultra $100.

# コード生成# AIエージェント# API・SDK# 生産性向上# 日本語対応

ゼットエーアイ

Z.ai

The official service for the GLM family of large language models developed by Z.ai (Zhipu AI). In chat (chat.z.ai), GLM-5.3-Flash can write, analyze data and build slides and websites. For developers, Z.ai offers an API (GLM-5.3 costs $1.4 input / $4.4 output per 1M tokens) and the "GLM Coding Plan", a flat-rate plan for using GLM in 20+ tools such as Claude Code and Cline (Lite $18, Pro $80, Max $168 per month). Model weights are published on Hugging Face. Officially supported model languages are English and Chinese, and the interface is in English.

# 生成AI（LLM）# AIチャットボット# コード生成# AIエージェント# API・SDK# テキスト生成

リムーブハンドライティング

RemoveHandwriting

An AI tool from Shenzhen Liwa Software Technology Co., Ltd. (Shenzhen, China) that removes only handwritten text, pen marks, and grading marks from images, PDFs, and scanned documents. It keeps printed questions, tables, and layout, so filled-in worksheets and tests can be used again. Offers a color-preserving Color model, an Ultra model for difficult pages, manual correction by expert staff, batch processing, and an API, and works in Japanese on the web, iPhone, and Android. New users get 3 free credits; paid plans are credit-based, starting at Basic $3.9/month (60 credits).

# 教育# PDF・文書生成# API・SDK# 業務効率化# 日本語対応

サクラノエーアイエンジン

さくらのAI Engine

SAKURA internet's inference API service that runs LLMs and other foundation models in data centers in Japan. Use gpt-oss-120b, Kimi, Qwen and more through OpenAI- and Anthropic-compatible APIs; text generation is free for up to 3,000 requests per month, with pay-as-you-go beyond that. A Private Edition that runs your chosen model on dedicated GPUs for a flat monthly fee (individually quoted) is also available.

# 生成AI（LLM）# API・SDK# 検索エンジン・RAG# 音声認識・音声合成# AIモデル構築・管理# 日本語対応

ランクレート

Runcrate

A GPU cloud for AI developers and AI agents from Aeonmind LLC. Rent 20 GPU models such as H100, A100 and RTX 4090 by the hour (most billed by the second), and call 223 models through an OpenAI-compatible inference API. Its MCP server lets Claude Code, Codex and Cursor launch GPUs. Pay as you go from a prepaid balance (from $20, no monthly fee).

# AIモデル構築・管理# API・SDK# 生成AI（LLM）# AIエージェント

マイクロソフト・ディシジョン・ワン

Microsoft-Decision-1

Microsoft-Decision-1 is a decision-specialized model that Microsoft released on October 9, 2026. Instead of generating text, it returns a probability for each fixed answer option for the text or JSON you pass in. It supports three question types — noul (probability that something is true), choice (one option from a set), and score (a value on an ordered scale) — and costs $0.042 per million input tokens with free output, the same as TypeSafe AI's Jev. Built by post-training Qwen3.5-9B, it is available on Microsoft Foundry (preview) and OpenRouter with a 32,768-token context. It suits high-volume repeated judgments such as ticket routing, prioritization, grading AI responses, and safety checks. Accuracy in Japanese has not been published.

# API・SDK# 検索エンジン・RAG# 自動化・省力化# コスト削減

エルエムスタジオ

LM Studio

A desktop app from US-based Element Labs, Inc. that lets you browse open models such as Qwen, Gemma, gpt-oss and DeepSeek on screen, download them to your own computer (Mac, Windows, Linux), and use them for chat or as an OpenAI-compatible local API server. It is free for both personal and work use, and the interface can be switched to Japanese. Its sister AI agent app, "LM Studio Bionic", also offers paid plans for using open models in the cloud (Bionic+ at $20/month, Pro at $100/month).

# 生成AI（LLM）# AIモデル構築・管理# API・SDK# AIエージェント# 日本語対応

オラマ

Ollama

An open-model runtime from US-based Ollama Inc. that lets you download and run open models such as Gemma, Qwen, DeepSeek and gpt-oss on your own computer (Mac, Windows, Linux). A single command, "ollama run", gets you started, and it also works as an OpenAI-compatible API server. Running locally is free and unlimited. Paid plans for running large models in the cloud are Pro at $20/month and Max at $100/month. The interface is in English.

# 生成AI（LLM）# AIモデル構築・管理# API・SDK

ハギングフェイス

Hugging Face

Machine learning platform run by Hugging Face (US) for publishing, sharing and finding AI models, datasets and demo apps (Spaces). With more than 2 million public models, it is the standard place to get open models such as Qwen and Stable Diffusion. Browsing and downloading are free. The inference API (Inference Providers) and GPU-backed Spaces are pay-as-you-go; PRO for individuals is $9/month and Team for organizations is $20/user/month. The interface is in English.

# AIモデル構築・管理# API・SDK# 生成AI（LLM）

コンフィーユーアイ

ComfyUI

Open-source (GPL-3.0) node-based generative AI tool developed by Comfy Org in the US. Build workflows by connecting nodes for image, video, audio and 3D models and processing steps, and run Stable Diffusion, FLUX, Wan and more on your own PC. Japanese UI available. The desktop app Comfy Desktop is free; the cloud version Comfy Cloud starts at $20/month for Standard ($16/month equivalent when billed yearly).

# 画像生成# 動画生成# 品質向上・クリエイティブ支援# API・SDK# 日本語対応

リムーブビージー

remove.bg

[Standalone site scheduled to close on December 1, 2026, moving to Canva] A background removal service that uses AI to detect the subject of an uploaded image, such as a person, product or animal, and cut out the background automatically (operated by Canva Austria GmbH, part of the Canva group). It is available on the web, with bulk editing, an API, desktop apps and a Photoshop plugin, but it will close at 9:00 CET on December 1, 2026, with background removal moving to Canva and the API to Leonardo.Ai. This page covers the shutdown schedule, refunds and credits, and where to migrate.

# 画像生成# デザイン生成# 小売・EC# API・SDK# 日本語対応

ウェイダブ

Weydub

Weydub is an AI localization tool for remaking existing videos and images for other languages. In one workspace it removes burned-in subtitles, logos and on-screen text without blurring, generates subtitles with speech recognition (ASR) and OCR, translates with AI, dubs with AI voices including voice cloning, and translates text inside images, with batch processing of up to 100 files and an API. It is the international brand built on the technology of the Chinese video tool GhostCut and is operated by PixelRock Limited in Hong Kong. A Japanese interface is available. Pricing is pay-as-you-go credit packs (from $10 for 200 credits, with an expiry period).

# 字幕生成# 動画生成# 要約・翻訳# 音声生成# API・SDK# 業務効率化# 日本語対応

Recommended Articles on API & SDK

What Jev Is [October 2026]: The AI That Generates No Text — How to Use It, Pricing, and How It Differs from Microsoft-Decision-1

On 9 October 2026, Microsoft released Microsoft-Decision-1, a decision model at the same price as Jev ($0.042 per million input tokens, free output) and with the same question types. This article explains what Jev — released by TypeSafe AI on 15 September 2026 — is: a System One model that generates no text and returns Choice (classification), Score (rating) and Noul (a 0–1 truth value) with probabilities; speculative fan-out, where stacking questions barely changes latency; how to get started from the waitlist to the Python SDK, HTTP API and Claude Code skill; routing low-confidence answers to people; and use cases. It then compares Jev with Microsoft-Decision-1 (access, context length, availability, Japanese support), explains which to choose and whether you can use them from Japan, and covers Jev's latest status in October (Series A, rate limits, OpenRouter listing). It also covers the official note that CJK accuracy is lower than English and the weak spots TypeSafe publishes itself.

[2026 Edition] 20 AI Tools Transforming Developer Workflows | Coding Assistance, Automated Code Review, No-Code Web Development, and Workflow Automation

A curated look at the best AI tools for developers in 2026. Covers 18 tools across coding assistance, automated code review, no-code development, and workflow automation — all aimed at improving productivity.

DeepSeek-V4.1-Flash-1017とは【2026年10月17日公開】APIの料金（円換算）・商用利用・安全性・V4-Proとの違い

DeepSeek（ディープシーク）は2026年10月17日、V4.1-Flashの小改良版「DeepSeek-V4.1-Flash-1017」を公開し、Web版・アプリ・API（deepseek-flash）を同日に切り替えました。重みはMITライセンスで商用利用でき、コード修正のベンチマークDeepSWE v1.1では77.0とClaude Opus 5（74.0）を上回りますが、数字は自己申告で、Opus 5が上の項目も多くあります。この記事では、公式のAPIドキュメントとHugging Faceのモデルカードにもとづいて、何が変わったか、APIの料金を円でいくらか（日本時間のピーク時間帯と半額のオフピーク）、11月1日から課金が始まるSearch／Fetch API、V4-Proの現状、中国のサーバーに保存されるデータの扱いと自社の環境で動かす選択肢を整理します。

Autonomous AI Agents Compared: 7 Picks for August 2026 — Grok Bot, Manus, Devin and How to Choose

You tried an AI agent and it ended at "wow, impressive" — because what it returned was an answer, not finished work. That changed in August 2026, when products that leave the work inside your actual tools arrived all at once. This article compares seven of them — Grok Bot, Hey Noah, Manus, Genspark AI, Devin, Kilo Code, and Cloudflare OS — along a single axis: where the output lands. It covers the unit of isolation, how much access you hand over, the three layers of cost, and where to start, all from primary sources.