modelbench日本語

Gemini 4 Argon

GoogleCurrent

Input price
$2.00
Per 1 million tokens (USD)
Output price
$10.00
Per 1 million tokens (USD)
Effort levels
3total
Default: Med
Benchmark results
37total
Benchmark types: 36

Model information

API model ID
gemini-4-argon
release date
2026-09-30
effort
Low / Med (default) / High
Effort source
Based on thinking level in Gemini API thinking documentation (aistudio.google.com/docs/thinking).
Claude Code
unavailable
Codex
unavailable
Availability notes
Support in Claude Code / Codex is not explicitly stated officially (Gemini API, AI Studio, Antigravity, etc. are official routes). The official model list for Claude Code / Codex (checked 2026-11-12) contains only Anthropic / OpenAI models, respectively, so both are false.
Availability source
https://aistudio.google.com/docs/models

Price (USD per 1 million tokens)

Prices do not vary by effort; token consumption does.

input$2.00
output$10.00
cache read$0.10
cache write—

Introductory price (through 2026-11-30). From 2026-12-01: input $4.00 / output $20.00 / cache $0.20. releaseDate is the announcement date (initially limited to the Fairwind Program); available to developers on 2026-10-13

Gemini 4 Argon gemini-4-argon Input $2.00 → $4.00 / Output $10.00 → $20.00 / Context caching $0.10 → $0.20

Source: https://aistudio.google.com/docs/pricing

Benchmark results

BenchmarkeffortScoreSource cost (metric)measured/reported bysourceConditions and extraction method
Terminal-Bench 4.0not statednot stated57.4%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not stated. Distinct from the AA measured value of 57.1% on the DeepMind Gemini pagetranscribed from table
Source excerptTerminal-bench 4.0 | 57.4%
Terminal-Bench 4.0HighHigh57.1%—Artificial AnalysisGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: a measured value from Artificial Analysis (labeled Terminal-Bench 4.0 (AA))transcribed from table
Source excerptTerminal-Bench 4.0: Gemini 4 Argon 57.1
Terminal-Bench-Science 0.1HighHigh52.9%—Artificial AnalysisGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: a measured value from Artificial Analysistranscribed from table
Source excerptTerminal-Bench Science 0.1: Gemini 4 Argon 52.9
FrontierSWE 2HighHigh55.0%—ProximalGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: values from the Proximal official leaderboardtranscribed from table
Source excerptFrontierSWE 2: Gemini 4 Argon 55.0
DeepSWE v1.1not statednot stated77.9%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptDeepSWE v1.1 | 77.9%
AutomationBenchHighHigh51.3%—ZapierGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: Zapier official leaderboard (private set)transcribed from table
Source excerptAutomationBench: Gemini 4 Argon 51.3
OSWorld 2.1 (offline set, partial)HighHigh74.7%—GoogleGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: self-measured. 82-task offline subset, maximum of 3 runs, partial scoretranscribed from table
Source excerptOSWorld-2.1 (offline subset): Gemini 4 Argon 74.7
OSWorld 2.0not statednot stated69.2%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptOSWorld-2.0 Offline subset, partial score | 69.2%
Agents' Last Exam v1.1(rubric score)HighHigh69.0%—GoogleGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: self-measured. 122 tasks, ALE-Claw harness, 5-hour window, averaged rubric Scoretranscribed from table
Source excerptAgents' Last Exam v1.1: Gemini 4 Argon 69.0
Agents' Last Exam (pass rate, version not stated)not statednot stated39.5%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not stated. The metric is Pass rate (distinct from rubric Score 69.0% in the Carbon comparison table); version not statedtranscribed from table
Source excerptAgent's Last Exam (Pass rate) | 39.5%
Chartography (without tools)HighHigh71.6%—not statedGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: without tools, following the Surge official leaderboardtranscribed from table
Source excerptChartography: Gemini 4 Argon 71.6
Artificial Analysis Intelligence Index v5.0HighHigh50—Artificial AnalysisGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: values reproduced from Artificial Analysis (third-party)transcribed from table
Source excerptArtificial Analysis Intelligence Index v5.0: Gemini 4 Argon 50
Vals Index v2.2HighHigh69.2%—Vals AIGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: values reproduced from Vals AItranscribed from table
Source excerptVals Index v2.2: Gemini 4 Argon 69.2
Vending-Bench 2HighHigh13,718—Andon LabsGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: Andon Labs official leaderboard (final balance $)transcribed from table
Source excerptVending-Bench 2: Gemini 4 Argon 13718
Blueprint-Bench 2HighHigh54.4%—Andon LabsGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: Andon Labs official leaderboard (similarity score)transcribed from table
Source excerptBlueprint-Bench 2: Gemini 4 Argon 54.4
GraphWalks BFS (256k to 1M)HighHigh84.2%—GoogleGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: self-measured. 200 questions, BFS F1transcribed from table
Source excerptGraphWalks BFS (256k-1M): Gemini 4 Argon 84.2
GraphWalks BFS (128k or less)not statednot stated99.7%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptGraphWalks Up to 128k, BFS (F1) | 99.7%
RiemannBenchHighHigh76.0%—Surge AIGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: Surge official leaderboardtranscribed from table
Source excerptRiemannBench: Gemini 4 Argon 76.0
CWE-bench v1HighHigh68.0%—Collinear AI / Artificial AnalysisGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: official leaderboard (independent evaluation by Collinear AI / Artificial Analysis), programmatic pass@1transcribed from table
Source excerptCWE-bench v1: Gemini 4 Argon 68.0
ExploitGymHighHigh42.0%—GoogleGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: self-measured. Antigravity harness, single trial, no internet, default 6-hour timeout (performance via Fairwind Program)transcribed from table
Source excerptExploitGym: Gemini 4 Argon 42.0
SEC-bench ProHighHigh83.4%—GoogleGemini 4 Carbon (Google DeepMind models page, benchmark table)official · Retrieved 2026-11-12Vendor-reported value. Google published table. methodology PDF: Gemini 4 Carbon/Argon used the highest Gemini API thinking setting, pass@1, and a single trial. Row-specific source: self-measured. 183 tasks (2026-05 JS engine split), mean Pass@1 over 3 trials, Antigravity harness (via Fairwind Program)transcribed from table
Source excerptSEC-bench Pro: Gemini 4 Argon 83.4
Vals Finance Agent v2not statednot stated65.4%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptVals Finance Agent v2 | 65.4%
Harvey Legal Agent Benchmarknot statednot stated19.6%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptHarvey's Legal Agent Benchmark | 19.6%
Vibe Code Benchnot statednot stated91.9%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptVibe Code Bench | 91.9%
PostTrainBenchnot statednot stated45.3%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptPostTrainBench | 45.3%
LABBench 2not statednot stated88.8%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptLABBench 2 | 88.8%
LVBenchnot statednot stated91.7%—GoogleGemini 4 Argon model cardofficial · Retrieved 2026-11-12Vendor-reported value. Self-reported by Google (model card, as of 2026-10). thinking setting not statedtranscribed from table
Source excerptLVBench | 91.7%
GPQA DiamondHighHigh96.4%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptgpqa_diamond.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.9640151515151515, stderr=0.011918129589602694, Started at=2026-10-13T17:15:08.000Z
FrontierMath Tiers 1-3(v2)HighHigh86.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptfrontiermath_tiers_1_3_v2.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.8596491228070176, stderr=0.020611471958164294, Started at=2026-10-13T17:14:57.000Z
FrontierMath Tier 4(v2)HighHigh78.1%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptfrontiermath_tier_4_v2.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.7804878048780488, stderr=0.06544589202438408, Started at=2026-10-13T17:14:57.000Z
OTIS Mock AIME 2024-2025HighHigh100.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptotis_mock_aime_2024_2025.csv: Model version=gemini-4-argon_high, Best score (across scorers)=1.0, stderr=0.0, Started at=2026-10-13T17:14:57.000Z
SimpleQA VerifiedHighHigh76.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptsimpleqa_verified.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.76, stderr=0.013512312258920845, Started at=2026-10-13T17:15:14.000Z
Chess Puzzles(Epoch AI)HighHigh59.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptchess_puzzles.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.59, stderr=0.04943110704237103, Started at=2026-10-13T17:15:18.000Z
MirrorCodeHighHigh58.9%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Gemini 4 Argon (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptmirrorcode.csv: Model version=gemini-4-argon_high, Best score (across scorers)=0.5888888888888889, stderr=0.1171381252539371, Started at=2026-10-14T10:00:34.000Z
Epoch Capabilities Index(ECI)not statednot stated166.48—Epoch AIEpoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)third-party · Retrieved 2026-11-12A per-model metric that combines results across multiple benchmarks and multiple efforts (not per effort). Calculated by Epoch AI.transcribed from table
Source excerpteci_scores.csv: Model=Gemini 4 Argon, eci=166.48, eci_ci_low=163.77, eci_ci_high=169.46
LMArena Text (style control, overall)HighHigh1,522—LMArenalmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)third-party · Retrieved 2026-11-12Human head-to-head votes (Bradley-Terry). Source label "Gemini 4 Argon (High)" (effort is in parentheses in the label). Release date 2026-11-09.transcribed from table
Source excerptmodel_name=Gemini 4 Argon (High), rating=1522.38, rating_lower=1516.9, rating_upper=1527.9, vote_count=19787, rank=2, category=overall, leaderboard_publish_date=2026-11-09
LMArena Text Coding (without style control)HighHigh1,540—LMArenalmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)third-party · Retrieved 2026-11-12Human head-to-head votes (Bradley-Terry). Source label "Gemini 4 Argon (High)" (effort is in parentheses in the label). Release date 2026-11-09.transcribed from table
Source excerptmodel_name=Gemini 4 Argon (High), rating=1540.03, rating_lower=1530.5, rating_upper=1549.6, vote_count=4614, rank=3, category=coding, leaderboard_publish_date=2026-11-09

Related rankings

References