Anthropic claims world’s best coding AI with Claude 4 Opus and Sonnet

Opus 4 can sustain almost a full work day of focused coding work, while Sonnet 4 is supposed to be excellent for thinking and reasoning.

Both models produce near-instant responses to queries, but can turn to reasoning and thinking for more demanding requests.

World’s best on coding?
Anthropic claims Opus is «the world’s best coding model,» and it edges out Gemini 2.5 Pro, o3 and GTP 4.1 on SWE-bench Verified, but cannot surpass OpenAI’s o3 on certain PhD-level benchmarks, according to TechCrunch.

Both models are «hybrid» models, according to Anthropics launch document, that can reason, think and perform agentic tasks, like browsing and searching your files or the internet to produce summaries of large data sets.

New memory features
Anthropic further says the new models «dramatically outperforms all previous models on memory capabilities», when given access to local files which can be used to store key information in the memory — for better «awareness, coherence and performance on agent tasks.»

Both have context windows of 1 million «tokens,» which translates to roughly 750,000 words, but differ in price.

Price details
In API access, Opus 4 comes with a price tag of $15 for input and $75 for output of a million tokens, while Sonnet 4 is less expensive at $3/$15 for the same data. This puts them at the higher end of the spectrum.

In comparison, the price for Gemini Pro 2.5 is $2.5/$15 and OpenAI’s o3 is $10/$40 for input/output.

Opus 4 is presently only available for Anthropic’s paid tiers, but free users get access to Sonnet 4.

Releasing more frequently
Going forward, Anthropic will switch to a more frequent release schedule for their models, according to a draft blog post seen by TechCrunch:

— We’re … shifting to more frequent model updates, delivering a steady stream of improvements that bring breakthrough capabilities to customers faster, it says, and continues — This approach keeps you at the cutting edge as we continuously refine and enhance our models.

The pervious flagship model from Anthopic, Claude 3.7 Sonnet, was released on February 24 this year.

Read more: Anthropics launch post includes some impressive video, an overview by TechCrunch, and the business angle from CNBC. The models also play Pokemon for 24 hours Wired notes.