本文へスキップ

AnthropicLLM

Claude Sonnet 5.5 API

Model IDanthropic/claude-sonnet-5.5

Claude Sonnet 5.5 for reasoning, coding and tool workflows through the Hermes chat API.

  • 1M context
  • ツール
  • Vision
  • Reasoning
  • JSON

プレイグラウンドで Claude Sonnet 5.5 を実行

Send a chat request with your account and see the exact token charge. Code needs an API key.

パラメーター

モデル
model
anthropic/claude-sonnet-5.5
messages[].content
messages[0].content
max_tokens
temperature

リクエストのプレビュー

{"model": "anthropic/claude-sonnet-5.5","messages": [],"stream": true,"max_tokens": 1024}

Web Playground はサインイン中のセッションでこのボディを送信します。コードからは同じボディを API キーとともに送信してください。

結果

Send a message to run Claude Sonnet 5.5 with your account.

Usage and charge appear here after a reply.

Claude Sonnet 5.5 の料金

USD per 1M tokens. The tier is chosen by the prompt size of each request.

料金を読み込み中…
  1. ReserveWhen a request is accepted, Hermes reserves credits for the estimated prompt plus max_tokens of output (8,192 when you set none). Set max_tokens to keep the reservation small.
  2. SettleWhen the response finishes, the reservation is replaced by the exact charge from the usage the model reports: prompt tokens at the input rate, cached prompt tokens at the cache-read rate, completion tokens at the output rate. The remainder is released immediately.
  3. FailuresIf the model rejects the request nothing is charged. If a stream is interrupted after tokens were generated, the generated text is estimated and charged.

Claude Sonnet 5.5 でできること

Anthropic positions Claude Sonnet 5.5 as a balance of speed and capability. Its documentation describes changes in thinking and tool behavior relative to Sonnet 5; integrations should test their actual route and response handling when migrating.

The Hermes catalog lists anthropic/claude-sonnet-5.5 with a 1,000,000-token context, vision, tools, reasoning, streaming and structured outputs. Anthropic's native platform controls and hosted tools are not automatically the same as the Hermes OpenAI-compatible route. Check the exact model's status and price before calling it.

ユースケース

  1. Coding assistance

    Analyze changes and use tool calls in an application-controlled workflow.

  2. Document and image analysis

    Combine text and supported image input; leave room for the response within the context limit.

  3. Structured workflows

    Request structured outputs and validate the result before use.

コードから Claude Sonnet 5.5 を呼び出す

Any OpenAI SDK works: set the base URL and your Hermes key.

  1. 1

    API キーを作成

    In the dashboard, with the chat:create scope. API キー
  2. 2

    Point your SDK at Hermes

    Base URL https://api.hermes-ai.net/api/v1
  3. 3

    Send a chat completion

    POST /chat/completions with anthropic/claude-sonnet-5.5

コード例

Set HERMES_API_KEY in your environment.

shell
export HERMES_API_KEY=your_api_key
Full chat API reference
curl https://api.hermes-ai.net/api/v1/chat/completions \
  -H "Authorization: Bearer $HERMES_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-sonnet-5.5",
    "messages": [{"role": "user", "content": "Say hello in one sentence."}],
    "max_tokens": 200
  }'

Request

POST https://api.hermes-ai.net/api/v1/chat/completions API key with the chat:create scope, or your signed-in session.

フィールド型備考
modelstringanthropic/claude-sonnet-5.5
messagesarrayOpenAI chat messages: system, user, assistant and tool roles; text and image_url parts.
streambooleanServer-sent events with a final usage chunk.
max_tokensintegerOutput cap; also sizes the credit reservation.
tools / tool_choicearray / objectFunction tools only.
response_formatobjectJSON mode and JSON schema where the model supports it.
temperature, top_p, stop, seed, reasoning_effortPassed through unchanged.

レスポンス

フィールド備考
choices[].message / deltaThe OpenAI chat completion shape.
usageprompt_tokens, completion_tokens and prompt_tokens_details.cached_tokens as reported by the model.
hermes.charged_microsExact charge in USD micro-units (JSON responses). Streams expose it on GET /api/v1/chat/requests.
x-hermes-requestRequest id header, also used in the billing ledger.
error.codeDocumented Hermes error code; branch on it, not on the message.

Claude Sonnet 5.5 API のよくある質問

Answers built from the model's catalog entry and the Hermes AI API contract.

Can I reuse a native Claude API integration unchanged?

Hermes exposes an OpenAI-compatible chat route. Adapt and test request fields, tool use and response parsing for this route instead of assuming native Claude API parameters are identical.

What is the Claude Sonnet 5.5 API on Hermes AI?

Claude Sonnet 5.5 by Anthropic is served through the OpenAI-compatible chat completions endpoint, https://api.hermes-ai.net/api/v1/chat/completions, with the model ID anthropic/claude-sonnet-5.5.

What is the context window of Claude Sonnet 5.5?

1,000,000 tokens, shared by the prompt and the output.

Does Claude Sonnet 5.5 support vision, tools and JSON output?

Supported: ツール, Vision, Reasoning および JSON. Not supported: オーディオ. Streaming works with stream: true.

How is Claude Sonnet 5.5 billed?

Per token, from prepaid credits. When a request is accepted, credits are reserved for the estimated prompt plus max_tokens of output (8,192 when you set none); when it finishes, the exact charge is taken from the usage the model reports and the rest is released. A rejected request is not charged. Rates are in the pricing table.

Can I use the OpenAI SDK with Claude Sonnet 5.5?

Yes. Set the base URL to https://api.hermes-ai.net/api/v1 and use your Hermes API key; send anthropic/claude-sonnet-5.5 as the model. Chat completions, streaming, tools and JSON output work unchanged.

What happens when I hit a rate limit?

Too many requests per minute return 429 RATE_LIMITED and too many in flight return 429 CONCURRENCY_LIMITED. If the model provider rate limits, you get 429 UPSTREAM_RATE_LIMITED and nothing is charged. Retry with backoff.

How do I get an API key for Claude Sonnet 5.5?

Create one under API キー in the dashboard with the chat:create scope, then add credits on the Billing page.