Claude Opus 5 API (claude-opus-5)

$4 / $20 per million tokens (input / output, effective) — one OpenAI-compatible endpoint at https://api.apimart.ai/v1, no monthly plan required. (observed 2026-09-17)

Get an API key · Live pricing · Model page · ⚡ 60-second quickstart

Why teams call Claude Opus 5 (claude-opus-5) through APIMart

• One key, entire catalog. The same https://api.apimart.ai/v1 base URL and Authorization header reach Claude Opus 5 (claude-opus-5) and 300+ other image, video and language models — switch the model field, not your client.
• $1 minimum, pay as you go. No subscription and no prepaid plan to size up front: top up from $1 and spend it on calls. There is no free quota to burn through first, so the price in this table is the price you pay.
• The charge comes back in the response. Every call reports the amount billed (cost / credits_cost), so a spend number is read per call instead of guessed at month end.
• Drop-in OpenAI shape. POST /v1/chat/completions with the same request body your client already sends; only base_url and model change.

Claude Opus 5 on APIMart is an Anthropic flagship route available through the OpenAI-compatible chat endpoint: per-million-token billing, prompt-cache write and read rates, and long-form output confirmed by real calls.

Endpoint: POST https://api.apimart.ai/v1/chat/completions (OpenAI-compatible). Streaming is the default — pass stream: false when you want one JSON object back.

The effective column is what you pay after the default group discount; data/model.json is refreshed daily by CI, and the usage block in every response tells you exactly which tokens were billed.

curl -sS https://api.apimart.ai/v1/chat/completions \
  -H "Authorization: Bearer $APIMART_API_KEY" -H 'Content-Type: application/json' \
  -d '{"model":"claude-opus-5","stream":false,"messages":[{"role":"user","content":"Name three retry rules."}]}'

import os, requests

BASE = "https://api.apimart.ai/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['APIMART_API_KEY']}", "Content-Type": "application/json"}

payload = requests.post(f"{BASE}/chat/completions", headers=HEADERS, timeout=120, json={
    "model": "claude-opus-5", "stream": False,
    "messages": [{"role": "user", "content": "Name three retry rules."}],
}).json()
print(payload["choices"][0]["message"]["content"])
print(payload["usage"])

Streaming, tool calling and JSON mode examples are in examples/ (t_curl.sh, python_chat.py).

These rows are actual completions recorded from this route, with the token usage the API returned and the cost computed from the effective rates above.

Full transcripts (including longer answers) are in data/samples.json.

What is the Claude Opus 5 API model id?

claude-opus-5. Sibling ids on the same endpoint include claude-opus-4-8, claude-opus-4-7 and claude-sonnet-4-6, so model selection is a string change.

How does prompt caching change the price?

Cached input is billed far below fresh input, while writing the cache costs a premium for 5-minute or 1-hour retention. For repeated system prompts the cached path is where the savings are.

Can I call it with the OpenAI SDK?

Yes for the chat-compatible path; the native Claude Messages shape is documented separately if you need Anthropic-specific fields.

Is long output billed differently?

Output tokens are billed per million like input, so a long answer is simply more output tokens — the usage block in the response tells you exactly what was consumed.

• claude opus 5 api
• claude opus 5 api pricing
• claude api
• anthropic api alternative
• llm api pricing comparison
• prompt caching api
• openai compatible api

Start with $1. Get an API key → check live pricing → open Claude Opus 5 (claude-opus-5) in the model library. The first call is three steps: submit, poll task_id, read the charged amount off the response.

Outbound APIMart links are minted through the promo link API; hand-made tracking parameters are rejected by tools/check_links.py in CI.

Claude Opus 5 is a third-party model served through APIMart; this repository publishes model ids, measured prices and real call outputs, and does not claim official status. Model names, prices and documentation belong to their respective owners. Endpoint reference: https://docs.apimart.ai/en/api-reference/texts/general/claude-messages.

README.md             model ids, token pricing, verified capabilities, real outputs
data/model.json       token rates for every tier (CI-refreshed)
data/samples.json     recorded completions with usage and computed cost
tools/snapshot.py     refresh pricing from the public payload
tools/check_links.py  attribution guard
examples/             curl and Python clients (streaming, tools, JSON mode)
.github/workflows/    daily price refresh + validation

MIT — see LICENSE.