Claude Sonnet 5.5
Anthropic’s fast mid-tier model, now within a whisker of Opus on most work.
About Claude Sonnet 5.5
Sonnet 5.5 is the Claude to try first for everyday coding and documents. It arrived at the end of September 2026 with the biggest jump Anthropic has shown this year: on Terminal-Bench 4.0 it went from 10.3% for Sonnet 5 to 70.6%, ahead of Opus 5.5’s 66.4%. On professional knowledge work it scored 1844 in Anthropic’s GDPval-AA test, two points behind Opus 5.5.
It sits a price tier below Opus, and Anthropic says it runs more than 30% faster than Sonnet 5. Thinking is on by default and can be dialled back. Where it gives way to Opus is open-ended work that needs sustained judgement over a long stretch — Anthropic says so itself.
Where it’s strong
Everyday coding, bug fixing and command-line agents.
Writing documents, slides and spreadsheets.
High-volume agent work where Opus is too slow or too costly.
A first choice for most new products.
Where to be careful
For long, open-ended work that needs judgement throughout, Anthropic says Opus 5.5 is clearly stronger.
Released on 28 September 2026, so there are fewer worked examples and community notes than for older models.
Some microbiology and virology requests may be wrongly flagged by Anthropic’s safety filters.
Price
For scale: 1,000 requests of about 3,000 tokens in and 1,000 out cost $18.40 — a long prompt and a page of answer each. Every rate is published and stays the same from one request to the next. See every model.
Specs
claude-sonnet-5-5Call it
Use the OpenAI SDK you already have — point it at Roar AI and name the model. Read the docs.
from openai import OpenAI
client = OpenAI(base_url="https://api.roar-ai.com/v1", api_key="roar_live_…")
reply = client.chat.completions.create(
model="claude-sonnet-5-5",
messages=[{"role": "user", "content": "Summarise this contract in plain English."}],
)Compare Claude Sonnet 5.5
Questions
How much does Claude Sonnet 5.5 cost?
Input is $2.30 per 1M tokens; output is $11.50 per 1M tokens. For scale, 1,000 requests of about 3,000 tokens in and 1,000 out cost $18.40. The rate is published and does not change from one request to the next.
Can I use Claude Sonnet 5.5 alongside other models?
Yes. Every model on Roar AI is on the same API key and the same monthly invoice, so switching from Claude Sonnet 5.5 to another model is a change to one word in your code.
Is Claude Sonnet 5.5 open source?
No. Anthropic does not publish the weights, so it is only available through an API like this one.