Claude Sonnet 5

Sonnet 5 gets you most of what Opus does for daily engineering at a lower price. Anthropic launched it in June 2026 and called it the most agentic Sonnet up to that point: it writes plans, operates browsers and terminals and keeps working on its own for longer. Anthropic puts its scores close to Claude Opus 4.8, with improvements over Sonnet 4.6 in reasoning, tool use, code and knowledge work.

On Kadraris the 1M-token window is both the default and the ceiling, with up to 128K output tokens.

What you can use it for

• Routine coding sessions in Claude Code, Cursor or Cline
• Tracking a bug in old, tangled code down to its actual cause
• Multi-step agents that call your tools and APIs in a loop
• Producing reports, emails and docs in bulk

Settings and limits

• Context window: 1,000,000 tokens; output up to 128K.
• Non-default temperature, top_p or top_k values return a 400 error. Leave them out.
• Adaptive thinking is enabled by default.
• Anthropic's new tokenizer turns the same text into about 30% more tokens than Sonnet 4.6 did, so compare the cost of full requests rather than rates alone.
• Use /v1/chat/completions or /v1/messages with the same sk-kd- key.

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-sonnet-5",
    messages=[{"role": "user",
               "content": "Add pytest cases for the edge "
                          "cases in parse_invoice()."}],
)
print(resp.choices[0].message.content)

Point it at your own function name and paste the code below the instruction.

Questions

How much is Sonnet 5 per million tokens?

$3.24 / $16.20 per 1M tokens on Kadraris. Keep the tokenizer change in mind when you compare with older Sonnet bills. A failed call costs $0.

Should I move to Sonnet 5.5?

Claude Sonnet 5.5 is the newer Sonnet and has a lower token rate here, but it rejects forced tool choice. Test your prompts on both before switching.

When is Opus worth it over Sonnet 5?

For large migrations or long unattended sessions, where one more mistake costs more than the price gap, try Claude Opus 5.5. The Opus 5.5 vs Sonnet 5 comparison helps you decide.

Claude Sonnet 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-sonnet-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-sonnet-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-sonnet-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-sonnet-5
claude