AI Model Release Tracker

Gemini 4 Carbon

Released

Gemini 4 Carbon is an AI model released by Google on Wednesday, Oct 28 2026, 15 days after Gemini 4 Argon Cyber Permissive. Benchmark results (shown below) cover Gray Swan IPI, SWE-sweep, FrontierSWE V2, Terminal-Bench-Science 0.1, RiemannBench, Agents’ Last Exam v1.1, and 12 more.

Announced API pricing

Announced introductory
InputWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it.
$2.00
Cached inputA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again.
$0.10
OutputWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one.
$10.00
  • Announced launch pricing; initial access is limited to selected cyber defenders and trusted Fairwind partners, with broader availability to follow.
  • After the introductory period, input is $4 and output is $20 per million tokens. Google did not specify an end date for Carbon's introductory period.
  • Cached input is discounted by 95% from the input rate. Cache-write pricing and thinking-token billing are not specified in the announcement.
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Verified October 28, 2026Read from blog.google. Click the check to open the source.

Benchmarks

Coding

SWE-sweepAgentic coding — subtask-weighted reward for software engineering with the mini-swe-agent harness
9.6%
FrontierSWE V2Ultra-long-horizon coding — Engineering problems that would occupy a person for days: systems implementation, performance work, scientific computing and AI research, with up to twenty hours per task. Partial credit is awarded, because finishing one outright is still rare. Higher is better.
61.2%
#1

Terminal & CLI

Terminal-Bench-Science 0.1Agentic scientific computing — The same command-line setup as Terminal-Bench, pointed at scientific work: the AI has to drive research tooling and computational workflows through to a result, rather than administer a machine. Version 0.1 is the first release of the task set, and scores run lower than on the general board. Higher is better.
64.8%
#1
Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better.
65.2%
#4 of 32

Agentic & tool use

Agents’ Last Exam v1.1Agentic computer use — averaged rubric score on the 122-task October 17, 2026 v1.1 set, using ALE-Claw with a five-hour limit; separate from earlier ALE scores
77.4%
CWE-bench v1Security patching — Collinear AI’s programmatically graded security-patch pass@1 rate; separate from exploit-generation benchmarks
71%
ExploitGym (causal pass rate)Cybersecurity — share of tasks producing a verified exploit of the intended vulnerability, with mitigations off; not a count of solved instances
54.8%10-hour budget
53.5%6-hour budget
10-hour budget
6-hour budget
SEC-bench Pro (JavaScript)Vulnerability discovery — pass@1 on the 183-task May 2026 JavaScript-engine split, averaged over three trials with the Antigravity harness
88.2%
OSWorld 2.1 (offline)Agentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.1, in its offline setting. Scores are not comparable with version 2.0. Higher is better.
81.5%
#1
AutomationBenchBusiness workflows — Tests whether the AI can run real multi-step business workflows — the kind of end-to-end office processes companies want to automate — from start to finish. Higher is better.
53.9%
#4 of 20

Reasoning & science

RiemannBenchAdvanced mathematics — mathematical problem-solving evaluation published by Surge
80%

Knowledge work

AA Intelligence Index v5.0Overall intelligence — Artificial Analysis composite intelligence index, version 5.0 with tools; separate from earlier index editions
59
Vals Index v2.2Professional work — Vals AI composite evaluation of professional tasks, version 2.2
70.4%
Vending-Bench 2 (USD)Business simulation — ending bank balance in US dollars in Andon Labs’ vending-business simulation; higher is better
16484

Multimodal

ChartographyChart tasks — The same chart-centred test with no tools: the AI has to read each chart unaided. Scores run far lower than the with-tools version, so read the two as separate tests. Higher is better.
72.5%no tools
#1
Blueprint-Bench 2Spatial reasoning — Can the AI reason about space and layout — for example, understanding a floor plan or blueprint? Higher is better.
58.6%
#1

Long context

GraphWalks BFS (256K–1M)Long context — breadth-first search F1 on the 200 GraphWalks problems with 256K to 1M tokens of context
96.2%

Robustness

Gray Swan IPIPrompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better.
0%k = 15
#1

Compare Gemini 4 Carbon with

Gemini 4 Carbon

Suggested comparisons

Gemini 4 CarbonvsGemini 4 ArgonGemini 4 CarbonvsGPT-6.1 TerraGemini 4 CarbonvsClaude Haiku 5.5Gemini 4 CarbonvsMuse Spark OpenGemini 4 CarbonvsGrok 4.7Gemini 4 CarbonvsDeepSeek-V4.1-Flash-1017Gemini 4 CarbonvsMistral Large 4Gemini 4 CarbonvsKimi K3.1 ProGemini 4 CarbonvsGLM-5.3-FlashGemini 4 CarbonvsQwen3.8-Max-0902Gemini 4 CarbonvsNemotron 3.5 Bolt E4B 4GB

About

Gemini 4 Carbon was announced on October 28, 2026 as Google's next frontier model after Gemini 4 Argon, with an emphasis on sustained software engineering and cybersecurity defense. Google described internal engineering runs lasting days, working through bugs across large codebases and low-level systems. Its launch evaluation listed 61.2% on FrontierSWE 2 and 65.2% on the Artificial Analysis run of Terminal-Bench 4.0, alongside evaluations of computer use, scientific work and multimodal understanding.

Google planned the initial rollout for selected cyber defenders and trusted partners through its Fairwind Program, while preparing broader developer, business and consumer access. That staged release paired gains in finding and repairing vulnerabilities with further work on model alignment, including fewer unsupported claims and attempts to take shortcuts in Google's internal testing. The external vulnerability-finding, exploitation and patching benchmarks were evaluated without guardrails, so their conditions should be distinguished from the safeguards and access restrictions of the rollout.

Frequently asked questions

Gemini 4 Carbon was released by Google on Wednesday, Oct 28 2026.

All Google releases

33 tracked