Gemini 4 Carbon
An internal Google checkpoint within the Gemini 4 model series, codenamed Carbon. It is described as a newer version than the Argon checkpoint, with reported coding capabilities comparable to those of Claude Opus 5.5.
Core Facts
Google is internally testing a new Gemini 4 checkpoint codenamed Carbon, whose coding capabilities are being benchmarked against Claude Opus 5.5.
Google DeepMind has released the next-generation frontier large model Gemini 4 Argon, positioned as a flagship model capable of sustained deep reasoning in complex, long-horizon professional workflows.
Google calls Gemini 4 Carbon a "new frontier model" and emphasizes its independent evaluation performance.
Timeline (last 90 days)
After Gemini 4 Carbon's rate of "fabricating instead of admitting it doesn't know" rose to 11%, that rate is still the second lowest among all models.
Google did not give a specific date for opening Gemini 4 Carbon to developers and ordinary users, only saying "as soon as possible".
Third-party testing shows that the proportion of Gemini 4 Carbon forcibly fabricating answers to questions it cannot answer rose from 4% to 11%.
After personally testing the two checkpoints Carbon B and Carbon Z, the uploader concluded that their coding performance falls far short of Claude Opus 5.5 and even shows almost no meaningful improvement over existing Gemini models.
Judging by past cadence, the interval from safety personnel to Pro user access is roughly two weeks or more.
Gemini 4 Carbon ranks first globally in SWE code-fixing capability, and in SWE2 software engineering capability it is second only to the flagship models of OpenAI and Anthropic.
Gemini 4 Carbon ranks first globally in TB terminal operations capability.
Gemini 4 Carbon ranks first globally in chart analysis (BB drawing comprehension).
Gemini 4 Carbon tops both AB automated operation and VB2 financial capability.
The RB high-speed reasoning test set for Gemini 4 Carbon includes higher-order mathematical reasoning related to the Riemann hypothesis, and its performance exceeds the flagship models of OpenAI and Anthropic, ranking first globally.
40 more timeline events
All Facts (20)
Google is internally testing a new Gemini 4 checkpoint codenamed Carbon, whose coding capabilities are being benchmarked against Claude Opus 5.5.
75%VerifiedGoogle DeepMind has released the next-generation frontier large model Gemini 4 Argon, positioned as a flagship model capable of sustained deep reasoning in complex, long-horizon professional workflows.
80%Partially VerifiedGoogle calls Gemini 4 Carbon a "new frontier model" and emphasizes its independent evaluation performance.
90%Partially VerifiedThe release of Gemini 4 Carbon comes only two weeks after the release of the previous generation, Gemini 4 Argon.
90%Partially VerifiedGoogle DeepMind officially released its next-generation frontier model Gemini 4 Carbon, which it calls its "strongest model."
85%Partially VerifiedGemini 4 Carbon is priced at $2 per million input tokens and $10 for output, with a 95% discount on cached input.
80%Partially VerifiedAccording to Artificial Analysis's evaluation, Gemini 4 Carbon ranks first on its intelligence index, just one point ahead of Cloud Opus 5.5.
80%Partially VerifiedGoogle officially states that Gemini 4 Carbon achieves significant improvements in coding, agentic tasks and cybersecurity.
80%Partially VerifiedGoogle's internal tests show that Gemini 4 Carbon's success rate in generating secure code is more than twice that of the previous generation Argon.
75%Partially VerifiedGemini 4 Carbon scored 59 on the Artificial Analysis Intelligence Index.
75%Partially VerifiedGoogle says Gemini 4 Carbon leads throughout the entire defensive workflow, from discovering vulnerabilities and proving they can be exploited to ultimately writing patches.
75%Partially VerifiedGemini 4 Carbon is currently unavailable to ordinary users, with Google prioritizing it for cyber defense defenders, distributed through the Fairwind program.
75%Partially VerifiedGemini 4 Carbon achieved a score of 71% on the CW Bench cybersecurity test.
65%Partially VerifiedGemini 4 Carbon's product description is: 'Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities.'
65%Partially VerifiedGemini 4 Carbon achieved 88.2% on the SEC-bench Pro benchmark.
65%Partially VerifiedAccess to Gemini 4 Carbon is first opened to vetted cyber defenders rather than being broadly released to the public.
65%Partially VerifiedIn the material currently crawled, no substantive content such as Gemini 4 Carbon's model parameters, context length, benchmark scores or pricing appears.
65%Partially VerifiedGemini 4 Carbon's per-task cost is about 72% lower than its competitors.
65%UnverifiedCarbon scored 54.8% on Exploit Gym, ranking first.
60%UnverifiedGemini 4 Carbon's cheating success rate on Terminal Bench was 0.0%.
60%