Menu Close
  • Home
  • News
  • Anthropic launches Claude Sonnet 5.5, a faster, lower-cost complement to Opus 5.5
    •

Anthropic launches Claude Sonnet 5.5, a faster, lower-cost complement to Opus 5.5

Aiden Reed 0 Comments
Illustration on a warm oatmeal background. Two curved wires run left to right toward a small jade finish post. The upper wire, labeled "PRIOR SONNET," is faint brown and crowded with many small, tightly packed beads that stop short of the post. The lower wire, labeled "NEW SONNET," is bright green with a few evenly spaced glossy beads, and its lead bead has already reached the post. It carries the tags "FEWER TOKENS" and "FASTER." A tag beside the post reads "SAME PRICE." At lower left, a chunky brown rotary knob labeled "EFFORT" has its pointer on "MEDIUM," with "LOW," "HIGH" and "MAX" around the rim. At lower right, the green wire forks and one red bead diverts onto a short branch ending in a red "CYBER SAFEGUARD" tag with a smaller "FALLBACK" tag.

Anthropic on Monday released Claude Sonnet 5.5, the second model in its Claude 5.5 family, the company announced. Anthropic said the model runs more than 30% faster than Claude Sonnet 5 and costs up to 30% less per task, and it positioned Sonnet 5.5 as a faster, lower-cost complement to Claude Opus 5.5, which it launched last week.

Pricing is unchanged from Sonnet 5 at $2 per million input tokens and $10 per million output tokens, with cache reads at $0.20 per million tokens, according to Anthropic. The company said the savings come from the model needing fewer tokens to do the same work. Opus 5.5 is priced at $4 per million input tokens and $20 per million output tokens.

Anthropic said Sonnet 5.5 is available now on all platforms, including Amazon Web Services, Google Cloud and Microsoft Azure, and developers can reach it on the Claude Platform as claude-sonnet-5-5. Developers who run Sonnet with thinking turned off will need to switch to a new between_tools setting before moving to the model, the company said.

Anthropic said Sonnet 5.5 is strongest at well-scoped everyday tasks, fixing bugs and creating documents, slides and spreadsheets. On Terminal-Bench 4.0, an agentic coding evaluation, it scored 70.6% compared with 10.3% for Sonnet 5, according to Anthropic. The company reported a CursorBench 4.0 score of 55.5%, against 57.8% for Opus 5.5, and a score of 1844 on GDPval-AA, a knowledge work test run by Artificial Analysis, against 1846 for Opus 5.5. Anthropic said Opus 5.5 “remains clearly stronger at complex, open-ended work requiring sustained judgment.”

“Sonnet is really for the cost-conscious customer where they might not need as much intelligence,” Theo Chu, a research product manager at Anthropic, told CNBC.

Anthropic said “Sonnet 5.5 doesn’t advance the frontier of our models’ capabilities,” so its alignment assessment focused on a targeted set of risks. Because the model’s cyber capabilities are a large improvement over Sonnet 5’s, it is the first Sonnet model to launch with safeguards similar to those on Opus 5.5, with higher-risk cybersecurity tasks falling back to Sonnet 5, the company said.

CNBC reported that the release is Anthropic’s second since CEO Dario Amodei called for a slowdown in AI development. Amodei earlier this month called on the global AI community to slow the pace of releasing new capabilities, Reuters reported, adding that the company is expanding its lineup ahead of a planned IPO. Anthropic said Claude Haiku 5.5, built for high-volume and cost-sensitive applications, will join the family in the coming weeks.

Sources

Tagged ,
0 0 votes
Article Rating
Subscribe
Notify of
0 Comments
Inline Feedbacks
View all comments
0
Would love your thoughts, please comment.x
()
x