本文へスキップ
125 model families from 19 providers

One API for image, video, music models and LLMs

Call every model with one API key and one prepaid balance. Media models run as tasks you submit and poll, chat speaks the OpenAI format, and you pay only for successful tasks.

  • OpenAI
  • Google
  • Anthropic
  • ByteDance
  • Qwen
  • Alibaba
  • xAI
  • Zhipu
  • Black Forest Labs
  • Kling
  • DeepSeek
  • MiniMax
  • Suno
  • Kunlun Wanwei
  • Meta
  • Moonshot
  • Vidu
  • Krea
  • PixVerse

One API layer for every model

Model access, keys, prepaid billing, task logs and agent tools in one place, behind one predictable interface.

01

Every output type

画像: 23, 動画: 24, 音楽・オーディオ: 4, LLM: 76

125 model families from 19 providers

02

One account, one interface

  • OpenAI-compatible chatPoint the OpenAI SDK at Hermes and keep your code.
  • One key, one prepaid balanceCredits shared by every image, video, music and chat model.
  • Async tasks with idempotencySubmit, poll and retry safely with an Idempotency-Key.
  • MCP server and Agent SkillClaude Code, Codex and Cursor call models for you.

OpenAI · Google · Anthropic · ByteDance · one key, one balance

03

Across the API lifecycle

Keys, 送信, ポーリング, Results, 請求, Logs, Support

Keys, tasks, usage, billing and support live in the same console.

MCP · Skills · llms.txt

One prompt connects your agent

Paste it into Claude Code, Codex, Cursor or any agent. It adds the MCP server, installs the Agent Skill and explains sign-in.

hermes-ai-api · setup prompt
Hermes AI API(https://api.hermes-ai.net)をセットアップして、画像・動画・音楽を生成できるようにしてください。

1. Hermes AI API の MCP サーバーを hermes-ai-api という名前で https://api.hermes-ai.net/mcp に追加してください(Streamable HTTP、OAuth サインイン)。
   - Claude Code:claude mcp add --transport http hermes-ai-api https://api.hermes-ai.net/mcp
   - Codex:codex mcp add hermes-ai-api --url https://api.hermes-ai.net/mcp
   - その他のクライアント:MCP 設定に追加してください。
2. Hermes AI API の Agent Skill をインストールしてください:https://api.hermes-ai.net/skills/hermes-ai-api.zip をダウンロードし、skills フォルダに展開します(Claude Code は ~/.claude/skills)。
3. ブラウザで Hermes AI API にサインインする方法を教えてください(Claude Code:/mcp を実行して hermes-ai-api を選択、Codex:codex mcp login hermes-ai-api)。その後、ツールが使えることを確認してください。

参考:https://api.hermes-ai.net/llms.txt
claude mcp add --transport http hermes-ai-api https://api.hermes-ai.net/mcp

Featured models

Featured families for image, video and LLM, with live prices.

View all models
  • Google

    Nano Banana Pro

    Google's Nano Banana Pro (Gemini 3.0 Pro Image) is an industry-leading cutting-edge text-to-image model, which can generate high-definition 4K images with excellent quality and has been fully optimized for mobile phone devices to ensure smooth operation. It is equipped with a ready-to-use REST inference API, has the best performance, zero coldstarts, and extremely affordable pricing.

    • テキストから画像
    • 画像から画像
    料金を読み込み中…
  • Google

    Nano Banana 2

    Lightweight text-to-image API built for high concurrency & instant response, with low-latency, budget-friendly image generation and editing. Perfect for scenarios with fast iteration, real-time preview or mass batch image creation. Low-price channel edition: much cheaper than official stable build, but unstable.

    • テキストから画像
    • 画像から画像
    料金を読み込み中…
  • OpenAI

    GPT Image 2.5

    GPT Image 2.5 Sunburst focuses on intricate detail, reliable typography, and controlled composition.

    • テキストから画像
    • 画像から画像
    料金を読み込み中…
  • Black Forest Labs

    FLUX 3

    Generate images from a text prompt.

    • テキストから画像
    • 画像から画像
    料金を読み込み中…
  • ByteDance

    Seedream 5.0

    Generate an image from a text prompt.

    • テキストから画像
    • 画像から画像
    料金を読み込み中…

Integrate in three steps

Create a key, point your code at Hermes and call any model the same way.

  1. 1

    API キーを作成

    Sign in and create a key in the console. One key works for every model.

  2. 2

    Point your code at Hermes

    Media models take one task request. For chat, set the OpenAI SDK base URL to https://api.hermes-ai.net/api/v1

  3. 3

    Call any model

    Submit a task, poll until it finishes and download the result. Charged only when it succeeds.

api.hermes-ai.net · POST /api/v1/jobs
# 1. Submit the task (asynchronous; returns 202 with a task id)
curl -X POST https://api.hermes-ai.net/api/v1/jobs \
  -H "Authorization: Bearer $HERMES_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
  "model": "nano-banana-pro-text-to-image-economy",
  "input": {
    "prompt": "A lighthouse on a rocky coast at sunrise, soft light",
    "resolution": "1k"
  }
}'

# 2. Poll until status is succeeded, failed or reconciliation_required
curl https://api.hermes-ai.net/api/v1/jobs/TASK_ID \
  -H "Authorization: Bearer $HERMES_API_KEY"
最小リクエスト例: nano-banana-pro-text-to-image-economy, スキーマから生成。

Why Hermes

Why developers build on Hermes

One platform, clear prices and an API that behaves the same for every model.

  • 01

    Pay only for successful tasks

    A quoted amount is reserved when a task is accepted, charged on success and returned if it fails.

  • 02

    One key, one balance

    Prepaid credits work across every image, video and music model and every LLM.

  • 03

    OpenAI-compatible

    Chat completions, streaming and tools through the OpenAI SDK with a new base URL.

  • 04

    Built for agents

    An MCP server with OAuth sign-in, an Agent Skill and llms.txt for coding agents.

  • 05

    Transparent per-model pricing

    Every model and variant shows its price beside the official one. 公式 API 価格より最大 28% お得.

  • 06

    6 languages

    The site reads in English, 中文, 日本語, 한국어, Español, Français.

FAQ

Common questions

All questions in the FAQ
  • 同期エンドポイントはありますか?

    Image, video and music models run as asynchronous tasks: POST /api/v1/jobs returns immediately with a task ID, and you poll GET /api/v1/jobs/{id} until a terminal status. LLMs are synchronous through POST /api/v1/chat/completions.

  • Can I use the OpenAI SDK?

    Yes. Set the base URL to https://api.hermes-ai.net/api/v1 and your Hermes key as the API key. Chat completions, streaming, tools and JSON output work unchanged; pick a model id from GET /api/v1/chat/models.

  • 送信中にネットワークが切れたらどうなりますか?

    同じ Idempotency-Key と同じボディで再試行してください。タスクが受け付けられていれば、二重に課金されることなくそのタスクが返されます(200、replayed: true)。

  • モデルを実行できないのはなぜですか?

    Check that the model is available and your balance covers the task price. The error code explains why a request cannot run.

  • クレジットを追加するには?

    Open Billing in the console to buy credits and view your balance, payments and usage. Credits become available after payment confirmation.

  • ローカルファイルを送信できますか?

    Upload them with POST /api/v1/media and put the returned media:<uuid> reference in the model's image, video or audio field. Public HTTPS URLs also work.

はじめに

Send your first task

Create a key, add credits and call any model in the catalog.