Google unveils Gemini 4 Carbon, first on the Artificial Analysis index with 59
The new model improves coding, agentic tasks and finding security flaws. Cyber defenders get it first; developers and consumers soon.
Google showed a new Gemini model that is better at coding and at finding security bugs. Security experts get it first, and everyone else soon.
Hype check
Louder than it deserves. The real change is smaller.
Our editorial opinion, based on the sources below. How we score
Your call: hype or real?
Real = it will matter in daily use. Hype = louder than it is. One vote, and it can’t be changed.
Our view and readers’ views are opinions, not statements of fact about a product.
Why it matters
It arrives only about two weeks after Argon and promises more accurate output with fewer unsupported claims. The figures are Google's own and need independent testing.
Watch: the numbers are Google's; we wait for independent tests.
The details
- Google announced Gemini 4 Carbon and says it ranks first on version 5.0 of the Artificial Analysis Intelligence Index with 59.
- On WebDev Arena it gained 89 Elo points to reach fourth place at 1769.
- In cybersecurity: 88.2% on SEC-bench Pro for finding flaws, 54.8% on ExploitGym, and 71.0% on CWE-bench for patching them.
- Google's internal tests show about half as many unsupported claims as Argon.
- It goes first to cyber defenders this week, then to developers, businesses and consumers "soon", with no date.
- Introductory price is $2 per million input and $10 for output, then $4 and $20.
Receipts
One trusted outlet reports it. Not confirmed yet. Open the sources and check us.
Written with the help of AI tools and checked against the sources above. How we work
See a mistake? Tell us and we’ll fix it.
