PL Log inCreate account
Launching in October 2026. Accounts, model catalog, calculator and docs are live. Studio is available in preview; generation is awaiting activation.

Chat & code Anthropic · 2026-06

Claude Sonnet 5 API

Anthropic's mid-tier model that plans, uses tools and codes close to Opus level

Input$3.24per 1M tokens
Output$16.20per 1M tokens

1,000 chatbot replies$11.341,500 input + 400 output tokens each

Cache reads on this model are billed at the input rate. Failed requests cost nothing.

Compare with other chat models →

Claude Sonnet 5 is Anthropic's Sonnet-class model released in June 2026. In Jantaris it has a 1,000,000-token context window, which is both its default and its maximum, and up to 128K output tokens. Anthropic calls it the most agentic Sonnet so far: it makes plans, works with browsers and terminals and runs on its own for longer stretches. Anthropic places its results close to Claude Opus 4.8 at a lower price, with gains over Sonnet 4.6 in reasoning, tool use, coding and knowledge work.

Best for

  • Day-to-day coding in Claude Code, Cursor or Cline
  • Debugging messy legacy code down to the root cause
  • Agents that call tools and APIs in multi-step loops
  • Drafting reports, emails and documentation at volume

How to call it

Swap the base URL in the OpenAI SDK and use the id claude-sonnet-5:

python
from openai import OpenAI

client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[{"role": "user", "content": "Write unit tests for utils/dates.py."}],
)
print(resp.choices[0].message.content)

It also works with the Anthropic SDK and Claude Code through /v1/messages, see the Claude Code guide.

FAQ

How much does Claude Sonnet 5 cost per 1M tokens?

In Jantaris: $3.24 / $16.20 per 1M tokens. Anthropic notes that Sonnet 5 uses a new tokenizer that produces about 30% more tokens for the same text than Sonnet 4.6, so compare the cost of whole requests, not just the rate per token.

Why does Sonnet 5 return a 400 error for my temperature setting?

Anthropic changed this in Sonnet 5: non-default values of temperature, top_p or top_k are rejected. Remove those parameters from the request. Adaptive thinking is also on by default.

Sonnet 5 or Opus 5.5 for Claude Code?

Sonnet 5 is the cheaper default for routine tasks. Switch to Claude Opus 5.5 for large migrations or long autonomous sessions where one extra mistake costs more than the price difference.

How to call Claude Sonnet 5 via API

from openai import OpenAI

client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[{"role": "user", "content": "Write a haiku about Baltic amber."}],
)
print(resp.choices[0].message.content)