AI Model Release Tracker

Gemini 4 Argon

Released

Gemini 4 Argon is an AI model released by Google on Wednesday, Sep 30 2026, 28 days after Gemini 3.8 Flash Cyber. It has a 1M token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover Next.js Evals, LAB-Bench 2, Finance Agent v2, LVBench, GraphWalks BFS F1 (up to 128K), BullshitBench v2, and 17 more.

API pricing

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
  • Output prices include reasoning tokens.
  • Introductory pricing through November 30, 2026. After that, $4 per million input tokens and $20 per million output tokens apply.
  • Publicly available on the paid Gemini API as gemini-4-argon since October 13, 2026. A cyber-permissive build is allowlist-only through the Fairwind Program.
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 19, 2026Read from ai.google.dev. Click the check to open the source.

Available from

Google AI Studio—flex$1.00$5.001M
Google AI Studio——$2.00$10.001M
Googleglobal—$2.00$10.001M
Googleus—$2.20$11.001M
Google AI Studio—priority$3.60$18.001M
Googleglobalpriority$3.60$18.001M
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.Checked November 15, 2026Third-party rates from openrouter.ai, not the lab’s own list price.

Benchmarks

Coding

Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better.via Next.js Evals
100%
#1
CursorBench 4.0Agentic coding — Cursor's own test of coding agents on ambiguous, multi-file tasks taken from real Cursor sessions — editing, refactoring, investigating a codebase, understanding what the user meant, managing jobs and following a design. Cursor runs each model at several reasoning efforts; each release here carries the score of its best listed effort. Scores aren't comparable with earlier CursorBench versions. Higher is better.via CursorBench
52.9%
#3 of 18
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better.
77.9%
#3 of 42
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.
55%
#7 of 11

Terminal & CLI

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.
57.4%
#9 of 36
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.
57.6%
#8 of 16

Agentic & tool use

RoboDojo SimRobotics — control of a simulated two-arm robot from camera images, average success rate
34.2%
#3 of 5
CWE-bench v1Security patching — Collinear AI’s programmatically graded security-patch pass@1 rate; separate from exploit-generation benchmarks
68%
#2 of 2
OSWorld 2.0Agentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.0 is a harder, refreshed task set. Higher is better.
69.2%
#6 of 12
Agent's Last ExamAgentic computer use — A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better.
39.5%pass@1
#7 of 15
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.
51.3%
#6 of 32

Reasoning & science

LAB-Bench 2Biology — Everyday tasks from a working biology lab — reading protocols, interpreting figures and sequence data, and answering the practical questions a researcher hits at the bench. Higher is better.
88.8%
#1
RiemannBenchAdvanced mathematics — mathematical problem-solving evaluation published by Surge
76%
#2 of 4
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better.via BenchLM
91.25%
#5 of 41

Finance

Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better.
65.4%
#1

Multimodal

LVBenchVideo understanding — Can the AI follow a very long video — up to an hour — and answer questions that need details from far apart in it? Higher is better.
91.7%
#1
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.
71.6%no tools
#3 of 15

Long context

GraphWalks BFS F1 (up to 128K)Long context — breadth-first search F1 on 650 problems with up to 128K tokens of context
99.7%
GraphWalks BFS (256K–1M)Long context — breadth-first search F1 on the 200 GraphWalks problems with 256K to 1M tokens of context
84.2%
#2 of 4

Robustness

BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.via BullshitBench
66%
#20 of 100
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.
0.7%k = 15
#2 of 15

Community preference

threejsevalCommunity preference (Three.js) — Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better.via threejseval
1714
#7 of 29

Source: CursorBench, retrieved 22 October 2026 · Source: BenchLM, retrieved 22 October 2026. Other sources are identified on the linked benchmark pages.

Compare Gemini 4 Argon with

Gemini 4 Argon

Suggested comparisons

Gemini 4 ArgonvsGemini 4 HydrogenGemini 4 ArgonvsGPT-6.1 TerraGemini 4 ArgonvsClaude Fable 5.2Gemini 4 ArgonvsMuse EmberGemini 4 ArgonvsGrok 4.7Gemini 4 ArgonvsDeepSeek-V4.1-Flash-1017Gemini 4 ArgonvsMistral Large 4Gemini 4 ArgonvsKimi K3.1 ProGemini 4 ArgonvsGLM-5.3-FlashGemini 4 ArgonvsQwen4-Max PreviewGemini 4 ArgonvsNemotron 3.5 Bolt E4B 4GB

About

Gemini 4 Argon was announced on September 30, 2026 with initial access for trusted defenders in Google's Fairwind Program. A wider rollout followed on October 13 through the paid Gemini API and selected Gemini app and Search experiences. Google positioned the model for sustained coding, research and professional workflows.

Argon's documented token limit was 1,048,576 shared across input and output. Long responses used a continuation mechanism beyond the default output allocation, so this was not simply a promise of one million output tokens in every request. Google emphasized keeping agents working through extended tool-driven tasks. Gemini 4 Carbon followed on October 28 with a further focus on sustained engineering and cybersecurity defense.

Frequently asked questions

Gemini 4 Argon was released by Google on Wednesday, Sep 30 2026.

All Google releases

34 tracked