Claude Opus 5.5

Opus 5.5 is the Claude I would start with for serious engineering work: changes that span a whole repository, agents that keep going for hours, and writing that has to respect a style guide. Anthropic released it in September 2026 as the first model of the Claude 5.5 family and says it matches Claude Fable 5.1 on most tasks while costing less to run.

Its training data runs to June 2026. On Kadraris it shares one key and balance with every other model in the catalog.

What you can use it for

• Migrations and refactors that touch hundreds of files in one plan
• Long Claude Code sessions where the agent edits, runs tests and retries on its own
• Loading a full repository, specification or set of contracts into one request
• Docs, release notes and technical articles that follow your house rules

Settings and limits

• Context window: 1,000,000 tokens; up to 128K output tokens per reply.
• Thinking stays on. There is no switch to disable it on Opus 5.5, so plan for reasoning tokens in the bill.
• Use /v1/chat/completions with the OpenAI SDK or /v1/messages with the Anthropic SDK.
• For Claude Code, set ANTHROPIC_BASE_URL to https://kadraris.com and put your sk-kd- key in ANTHROPIC_AUTH_TOKEN.
• Agent sessions fire many requests. Create a separate key with its own monthly limit for them.

Example request

from openai import OpenAI

client = OpenAI(
    base_url="https://kadraris.com/v1",
    api_key="sk-kd-...",
)
resp = client.chat.completions.create(
    model="claude-opus-5.5",
    messages=[{"role": "user",
               "content": "Outline the steps to move this "
                          "Django app to async views."}],
)
print(resp.choices[0].message.content)

Change the framework names to match your stack. More setup notes sit in the Claude Code guide.

Questions

How is Opus 5.5 billed?

Per token at $3.78 / $18.90 per 1M tokens, taken from a prepaid balance with no subscription and no monthly minimum. The balance does not expire, and a call that errors out costs $0.

Opus 5.5 or Sonnet 5 for a coding agent?

Use Claude Sonnet 5 for routine edits and Opus 5.5 for large, risky changes where one wrong step costs more than the rate difference. The Opus 5.5 vs Sonnet 5 page lays out both.

When should I pay extra for Fable 5.1?

Only after Opus 5.5 has failed your own tests on a task. Anthropic positions the two at a similar level for most work, so the higher tier is a fallback, not a default.

Claude Opus 5.5 in your code

from openai import OpenAI

client = OpenAI(base_url="https://kadraris.com/v1", api_key="sk-kd-...")
resp = client.chat.completions.create(
    model="claude-opus-5.5",
    messages=[{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}],
)
print(resp.choices[0].message.content)

curl https://kadraris.com/v1/chat/completions \
  -H "Authorization: Bearer $KADRARIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "claude-opus-5.5", "messages": [{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}]}'

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://kadraris.com/v1", apiKey: process.env.KADRARIS_API_KEY });
const resp = await client.chat.completions.create({
  model: "claude-opus-5.5",
  messages: [{ role: "user", content: "Suggest three opening shots for a short film about a cafe." }],
});
console.log(resp.choices[0].message.content);

export ANTHROPIC_BASE_URL=https://kadraris.com
export ANTHROPIC_AUTH_TOKEN=sk-kd-...
export ANTHROPIC_MODEL=claude-opus-5.5
claude