As of October 20, 2026, Google Gemini 4 Argon ranks #5 of 213 models overall with 52.6 on the Artificial Analysis Intelligence Index. Its strongest category is reasoning, where it ranks #1 of 203. Google Gemini 4 Argon costs $2.00 per million input tokens and $10.00 per million output tokens. It was released on October 13, 2026.
Its output price ranks 282 of 349 priced models, counting from the cheapest.
Listed on OpenRouter October 13, 2026, 7 days before our data. Results so far on 91 of 190 public boards; this page updates every day as boards publish.
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...
google/gemini-4-argon The tick on a coding bar is the median of every model measured on it. Outside coding, the place is this model's rank on that board. A board that never ran the model says so.
Measured in 4 of 6 categories. A place counts only the models measured on that board.
The index averages only the coding benchmarks that have a result; the others are skipped, never filled in. How it is weighted
Not measured on SWE-bench Lite, SWE-bench (full test set), SWE-bench Multimodal, SWE-bench Multilingual, AA Coding Index, LiveCodeBench (Vals AI), SWE-bench Verified (Vals AI), SkillsBench (Vals AI), LiveCodeBench (Artificial Analysis run), Terminal-Bench Hard (Artificial Analysis run), Terminal-Bench, DeepSWE, GSO-Bench, WeirdML, MirrorCode, Cybench, LMArena WebDev: from an image
Not measured on AA Agentic Index, Design Arena: agents building web apps, Design Arena: agents building full-stack apps, Design Arena: agents building mobile apps, Design Arena: agents building Android apps, Design Arena: agents building Godot games, Design Arena: agents building games, Design Arena: HTML slides, Design Arena: PowerPoint slides, Design Arena: spreadsheets, τ²-Bench Telecom, METR time horizons, Remote Labor Index, GDPval, OSWorld 2.0, BALROG, DeepResearch Bench, PostTrainBench, TheAgentCompany, OSWorld, SpireClimb, LMArena Search
Not measured on GPQA Diamond, MMMU Pro (Vals AI), GPQA Diamond (Artificial Analysis run), Humanity's Last Exam, SimpleBench, VPCT, BIG-Bench Hard, LMArena Vision: diagrams, LMArena Vision: homework
Not measured on AIME, MATH 500 (Vals AI), MGSM (Vals AI), AA Math Index, AIME 2025 (Artificial Analysis run), MATH level 5, GSM8K
Not measured on MMLU Pro, CaseLaw (Vals AI), CorpFin (Vals AI), TaxEval (Vals AI), MortgageTax (Vals AI), MedQA (Vals AI), MMLU Pro (Artificial Analysis run), MMLU, GeoBench, AI2 Reasoning Challenge
Not measured on Design Arena: SVG, Design Arena: ASCII art
Not measured on Vals Multimodal Index, IFBench, Fiction.LiveBench, Creative writing (Lech Mazur), CL-bench, Surface Evolver Bench, ExploitBench, HellaSwag, WinoGrande, PIQA, CL-bench Life, LMArena Vision, LMArena Text: German, LMArena Text: Japanese, LMArena Text: Korean, LMArena Text: French, LMArena Text: Polish, LMArena Vision: English, LMArena Vision: Chinese, LMArena Vision: reading text, LMArena Vision: humor, LMArena Vision: captioning, LMArena Vision: recognizing things, LMArena Vision: creative writing, LMArena Document
Not measured on Design Arena: images, Design Arena: graphic design, Design Arena: logos, LMArena Text to Image, LMArena Image Edit
Public boards shown as published (Vals AI, Artificial Analysis, Design Arena) and kept out of the coding index. Each links to its source on the benchmark page.
| Benchmark | Result | Index weight |
|---|---|---|
| Aider polyglotCoding | Not measured | 35% |
| SWE-bench VerifiedCoding | Not measured | 30% |
| SWE-bench LiteCoding | Not measured | Not in index |
| SWE-bench (full test set)Coding | Not measured | Not in index |
| SWE-bench MultimodalCoding | Not measured | Not in index |
| SWE-bench MultilingualCoding | Not measured | Not in index |
| LiveBench codingCoding | 89.7% | 20% |
| LMArena WebDevCoding | 1678 | 15% |
| Vals IndexGeneral | 69.2% | Not in index |
| GPQA DiamondReasoning | Not measured | Not in index |
| AIMEMath | Not measured | Not in index |
| MMLU ProKnowledge | Not measured | Not in index |
| AA Intelligence IndexIntelligence | 52.6 | Not in index |
| AA Coding IndexCoding | Not measured | Not in index |
| AA Agentic IndexAgentic work | Not measured | Not in index |
| Design Arena: websitesWeb design | 1315 | Not in index |
| Design Arena: UI componentsWeb design | 1333 | Not in index |
| Design Arena: game devWeb design | 1372 | Not in index |
| Design Arena: data vizWeb design | 1336 | Not in index |
| Design Arena: 3DWeb design | 1389 | Not in index |
| Design Arena: all code categoriesWeb design | 1328 | Not in index |
| Design Arena: SVGWeb design | Not measured | Not in index |
| Design Arena: ASCII artWeb design | Not measured | Not in index |
| Design Arena: agents building web appsAgentic work | Not measured | Not in index |
| Design Arena: agents building full-stack appsAgentic work | Not measured | Not in index |
| Design Arena: agents building mobile appsAgentic work | Not measured | Not in index |
| Design Arena: agents building Android appsAgentic work | Not measured | Not in index |
| Design Arena: agents building Godot gamesAgentic work | Not measured | Not in index |
| Design Arena: agents building gamesAgentic work | Not measured | Not in index |
| Design Arena: HTML slidesAgentic work | Not measured | Not in index |
| Design Arena: PowerPoint slidesAgentic work | Not measured | Not in index |
| Design Arena: spreadsheetsAgentic work | Not measured | Not in index |
| Design Arena: imagesImage generation | Not measured | Not in index |
| Design Arena: graphic designImage generation | Not measured | Not in index |
| Design Arena: logosImage generation | Not measured | Not in index |
| LiveBench reasoningReasoning | 95.1% | Not in index |
| LiveBench data analysisGeneral | 85.2% | Not in index |
| LiveBench researchAgentic work | 83.1% | Not in index |
| LiveBench browserAgentic work | 92.0% | Not in index |
| LiveBench long agenticAgentic work | 69.9% | Not in index |
| LiveBench designWeb design | 73.7% | Not in index |
| LiveCodeBench (Vals AI)Coding | Not measured | Not in index |
| SWE-bench Verified (Vals AI)Coding | Not measured | Not in index |
| Terminal-Bench 4 (Vals AI)Coding | 57.6% | Not in index |
| Vibe Code Bench (Vals AI)Coding | 91.9% | Not in index |
| Code migration (Vals AI)Coding | 68.2% | Not in index |
| ProgramBench (Vals AI)Coding | 2.5% | Not in index |
| SREBench (Vals AI)Agentic work | 44.3% | Not in index |
| IOI (Vals AI)Coding | 100.0% | Not in index |
| Terminal-Bench 2.1 (Vals AI)Coding | 57.7% | Not in index |
| SWE-bench Multimodal (Vals AI)Coding | 56.0% | Not in index |
| Vibe Code Bench 1-100 (Vals AI)Coding | 21.2% | Not in index |
| SkillsBench (Vals AI)Coding | Not measured | Not in index |
| CyberBench (Vals AI)Coding | 77.9% | Not in index |
| Finance Agent (Vals AI)Agentic work | 65.4% | Not in index |
| Tax Agent Bench (Vals AI)Agentic work | 44.6% | Not in index |
| Legal Research Bench (Vals AI)Agentic work | 54.8% | Not in index |
| Harvey's Legal Agent Benchmark (Vals AI)Agentic work | 27.1% | Not in index |
| Excel Modeling Benchmark (Vals AI)Agentic work | 75.2% | Not in index |
| BioMysteryBench (Vals AI)Agentic work | 76.3% | Not in index |
| Terminal-Bench-Science (Vals AI)Agentic work | 44.3% | Not in index |
| Vals Science IndexAgentic work | 55.4% | Not in index |
| Time Horizon Index (Vals AI)Agentic work | 52.5% | Not in index |
| CUA-bench (Vals AI)Agentic work | 4.8% | Not in index |
| Robo-bench (Vals AI)Agentic work | 0.8% | Not in index |
| Public Benefits Bench (Vals AI)General | 69.8% | Not in index |
| MysteryMechanism (Vals AI)Reasoning | 45.5% | Not in index |
| MMMU Pro (Vals AI)Reasoning | Not measured | Not in index |
| MATH 500 (Vals AI)Math | Not measured | Not in index |
| MGSM (Vals AI)Math | Not measured | Not in index |
| SAGE (Vals AI)General | 53.6% | Not in index |
| LegalBench (Vals AI)Knowledge | 88.3% | Not in index |
| CaseLaw (Vals AI)Knowledge | Not measured | Not in index |
| CorpFin (Vals AI)Knowledge | Not measured | Not in index |
| TaxEval (Vals AI)Knowledge | Not measured | Not in index |
| MortgageTax (Vals AI)Knowledge | Not measured | Not in index |
| MedQA (Vals AI)Knowledge | Not measured | Not in index |
| MedCode (Vals AI)Knowledge | 58.8% | Not in index |
| MedScribe (Vals AI)General | 87.4% | Not in index |
| Vals Multimodal IndexGeneral | Not measured | Not in index |
| AA Math IndexMath | Not measured | Not in index |
| GPQA Diamond (Artificial Analysis run)Reasoning | Not measured | Not in index |
| Humanity's Last Exam (Artificial Analysis run)Reasoning | 57.1% | Not in index |
| MMLU Pro (Artificial Analysis run)Knowledge | Not measured | Not in index |
| LiveCodeBench (Artificial Analysis run)Coding | Not measured | Not in index |
| SciCodeCoding | 61.8% | Not in index |
| Terminal-Bench Hard (Artificial Analysis run)Coding | Not measured | Not in index |
| IFBenchGeneral | Not measured | Not in index |
| Long context reasoning (AA-LCR)General | 79.7% | Not in index |
| τ²-Bench TelecomAgentic work | Not measured | Not in index |
| AIME 2025 (Artificial Analysis run)Math | Not measured | Not in index |
| Terminal-BenchCoding | Not measured | Not in index |
| DeepSWECoding | Not measured | Not in index |
| FrontierCodeCoding | 47.5% | Not in index |
| FrontierSWECoding | 55.0% | Not in index |
| GSO-BenchCoding | Not measured | Not in index |
| WeirdMLCoding | Not measured | Not in index |
| MirrorCodeCoding | Not measured | Not in index |
| APEX-AgentsAgentic work | 82.4% | Not in index |
| METR time horizonsAgentic work | Not measured | Not in index |
| Remote Labor IndexAgentic work | Not measured | Not in index |
| GDPvalAgentic work | Not measured | Not in index |
| OSWorld 2.0Agentic work | Not measured | Not in index |
| BALROGAgentic work | Not measured | Not in index |
| DeepResearch BenchAgentic work | Not measured | Not in index |
| PostTrainBenchAgentic work | Not measured | Not in index |
| GPQA Diamond (Epoch AI run)Reasoning | 96.4% | Not in index |
| ARC-AGIReasoning | 98.5% | Not in index |
| ARC-AGI-2Reasoning | 91.3% | Not in index |
| Humanity's Last ExamReasoning | Not measured | Not in index |
| SimpleBenchReasoning | Not measured | Not in index |
| Chess puzzlesReasoning | 59.0% | Not in index |
| Mystery game puzzlesReasoning | 64.0% | Not in index |
| EBR-benchReasoning | 64.8% | Not in index |
| OTIS Mock AIMEMath | 100.0% | Not in index |
| FrontierMath, tiers 1 to 3Math | 86.0% | Not in index |
| FrontierMath, tier 4Math | 78.0% | Not in index |
| ProofBenchMath | 99.0% | Not in index |
| MATH level 5Math | Not measured | Not in index |
| GSM8KMath | Not measured | Not in index |
| SimpleQA VerifiedKnowledge | 76.0% | Not in index |
| MMLUKnowledge | Not measured | Not in index |
| GeoBenchKnowledge | Not measured | Not in index |
| DTBenchGeneral | 97.6% | Not in index |
| LMCAGeneral | 63.7% | Not in index |
| Fiction.LiveBenchGeneral | Not measured | Not in index |
| Creative writing (Lech Mazur)General | Not measured | Not in index |
| CL-benchGeneral | Not measured | Not in index |
| Surface Evolver BenchGeneral | Not measured | Not in index |
| Furniture assemblyGeneral | 72.5% | Not in index |
| ExploitBenchGeneral | Not measured | Not in index |
| CybenchCoding | Not measured | Not in index |
| TheAgentCompanyAgentic work | Not measured | Not in index |
| OSWorldAgentic work | Not measured | Not in index |
| SpireClimbAgentic work | Not measured | Not in index |
| VPCTReasoning | Not measured | Not in index |
| BIG-Bench HardReasoning | Not measured | Not in index |
| AI2 Reasoning ChallengeKnowledge | Not measured | Not in index |
| HellaSwagGeneral | Not measured | Not in index |
| WinoGrandeGeneral | Not measured | Not in index |
| PIQAGeneral | Not measured | Not in index |
| CL-bench LifeGeneral | Not measured | Not in index |
| ALE-BenchCoding | 2100.1 | Not in index |
| Vending-Bench 2Agentic work | $13,718 | Not in index |
| Epoch Capabilities IndexIntelligence | 164.6 | Not in index |
| LMArena TextGeneral | 1532 | Not in index |
| LMArena Text: codingCoding | 1547 | Not in index |
| LMArena Text: hard promptsReasoning | 1546 | Not in index |
| LMArena Text: mathMath | 1536 | Not in index |
| LMArena Text: instruction followingGeneral | 1536 | Not in index |
| LMArena Text: creative writingGeneral | 1527 | Not in index |
| LMArena VisionGeneral | Not measured | Not in index |
| LMArena Text: EnglishGeneral | 1538 | Not in index |
| LMArena Text: non-EnglishGeneral | 1522 | Not in index |
| LMArena Text: ChineseGeneral | 1608 | Not in index |
| LMArena Text: GermanGeneral | Not measured | Not in index |
| LMArena Text: JapaneseGeneral | Not measured | Not in index |
| LMArena Text: KoreanGeneral | Not measured | Not in index |
| LMArena Text: FrenchGeneral | Not measured | Not in index |
| LMArena Text: SpanishGeneral | 1504 | Not in index |
| LMArena Text: RussianGeneral | 1534 | Not in index |
| LMArena Text: PolishGeneral | Not measured | Not in index |
| LMArena Text: multi-turnGeneral | 1556 | Not in index |
| LMArena Text: longer queriesGeneral | 1546 | Not in index |
| LMArena Text: expertReasoning | 1551 | Not in index |
| LMArena Text: hard prompts in EnglishReasoning | 1548 | Not in index |
| LMArena Text: software and ITCoding | 1556 | Not in index |
| LMArena Text: mathematical workMath | 1556 | Not in index |
| LMArena Text: scienceKnowledge | 1548 | Not in index |
| LMArena Text: medicine and healthcareKnowledge | 1554 | Not in index |
| LMArena Text: legal and governmentKnowledge | 1549 | Not in index |
| LMArena Text: business and financeKnowledge | 1525 | Not in index |
| LMArena Text: writing and languageGeneral | 1529 | Not in index |
| LMArena Text: entertainment and mediaGeneral | 1526 | Not in index |
| LMArena WebDev: ReactCoding | 1692 | Not in index |
| LMArena WebDev: HTMLCoding | 1643 | Not in index |
| LMArena WebDev: from an imageCoding | Not measured | Not in index |
| LMArena Vision: EnglishGeneral | Not measured | Not in index |
| LMArena Vision: ChineseGeneral | Not measured | Not in index |
| LMArena Vision: reading textGeneral | Not measured | Not in index |
| LMArena Vision: diagramsReasoning | Not measured | Not in index |
| LMArena Vision: homeworkReasoning | Not measured | Not in index |
| LMArena Vision: humorGeneral | Not measured | Not in index |
| LMArena Vision: captioningGeneral | Not measured | Not in index |
| LMArena Vision: recognizing thingsGeneral | Not measured | Not in index |
| LMArena Vision: creative writingGeneral | Not measured | Not in index |
| LMArena SearchAgentic work | Not measured | Not in index |
| LMArena DocumentGeneral | Not measured | Not in index |
| LMArena Text to ImageImage generation | Not measured | Not in index |
| LMArena Image EditImage generation | Not measured | Not in index |
A list price says what a token costs. The cost per solved task says what a working answer costs, which is the number that decides.
Neither board that publishes what a SWE-bench Verified attempt costs, Vals AI and swebench.com, has run this model. We leave the number out rather than estimate it.
LiveBench, on its own coding tasks, publishes what one task cost this model: $5.076, at a score of 89.7%. LiveBench counts that as $5.659 per successful task. Different tasks from SWE-bench, so it is not our number and does not enter its ranking. LiveBench coding
Beside a result, many boards publish what the run cost, how long it took, the tokens it spent or the votes behind a rating. Each figure as its board printed it, one board per row.
| Board | Result | What the run cost | Also published |
|---|---|---|---|
| LiveBench coding | 89.7% | $5.0763 per task | Output tokens per task 115,041 |
| Vals Index | 69.2% | $13.37 per test | Duration, as printed 44m 58s |
| LiveBench reasoning | 95.1% | $0.0781 per task | Output tokens per task 7,571 |
| LiveBench data analysis | 85.2% | $0.2843 per task | Output tokens per task 26,732 |
| LiveBench research | 83.1% | $5.9556 per task | Output tokens per task 144,282 |
| LiveBench browser | 92.0% | $3.4339 per task | Output tokens per task 15,322 |
| LiveBench long agentic | 69.9% | $13.8419 per task | Output tokens per task 327,753 |
| LiveBench design | 73.7% | $1.1604 per task | Output tokens per task 60,890 |
| Terminal-Bench 4 (Vals AI) | 57.6% | $17.64 per test | Duration, as printed 53m 35s |
| Vibe Code Bench (Vals AI) | 91.9% | $8.11 per test | Duration, as printed 37m 28s |
| Code migration (Vals AI) | 68.2% | $57.82 per test | Duration, as printed 2h 12m |
| ProgramBench (Vals AI) | 2.5% | $17.66 per test | Duration, as printed 1h 11m |
| SREBench (Vals AI) | 44.3% | $30.45 per test | Duration, as printed 1h 35m |
|---|---|---|---|
| IOI (Vals AI) | 100.0% | $5.31 per test | Duration, as printed 24m 19s |
| SWE-bench Multimodal (Vals AI) | 56.0% | $11.40 per test | Duration, as printed 33m 38s |
| Vibe Code Bench 1-100 (Vals AI) | 21.2% | $18.66 per test | Duration, as printed 1h 43m |
| CyberBench (Vals AI) | 77.9% | $5.11 per test | Duration, as printed 19m 10s |
| Finance Agent (Vals AI) | 65.4% | $4.38 per test | Duration, as printed 16m 21s |
| Tax Agent Bench (Vals AI) | 44.6% | $3.47 per test | Duration, as printed 16m 26s |
| Legal Research Bench (Vals AI) | 54.8% | $6.45 per test | Duration, as printed 31m 21s |
| Harvey's Legal Agent Benchmark (Vals AI) | 27.1% | $10.55 per test | Duration, as printed 29m 15s |
| Excel Modeling Benchmark (Vals AI) | 75.2% | $11.80 per test | Duration, as printed 38m 26s |
| BioMysteryBench (Vals AI) | 76.3% | $5.71 per test | Duration, as printed 39m 57s |
| Terminal-Bench-Science (Vals AI) | 44.3% | $44.90 per test | Duration, as printed 1h 46m |
| Vals Science Index | 55.4% | $18.40 per test | Duration, as printed 57m 00s |
| Public Benefits Bench (Vals AI) | 69.8% | $2.79 per test | Nothing else |
| MysteryMechanism (Vals AI) | 45.5% | $4.59 per test | Duration, as printed 24m 55s |
| FrontierSWE | 55.0% | $129.36 average cost per task | Best of 5 runs 64.3% · Hours per task 10.6 h |
| ProofBench | 99.0% | $0.783 per test | Time per proof 6m 20s |
| ALE-Bench | 2100.1 | $1.9497 per problem | Tokens per problem 192,623 |
| LMArena WebDev | 1678 Elo | Not published | Votes 1,728 · Interval (±) ±15.6 |
| Design Arena: websites | 1315 Elo | Not published | Battles 7,702 |
| Design Arena: UI components | 1333 Elo | Not published | Battles 542 |
| Design Arena: game dev | 1372 Elo | Not published | Battles 1,209 |
| Design Arena: data viz | 1336 Elo | Not published | Battles 287 |
| Design Arena: 3D | 1389 Elo | Not published | Battles 1,071 |
| Design Arena: all code categories | 1328 Elo | Not published | Battles 10,285 |
| Terminal-Bench 2.1 (Vals AI) | 57.7% | Not published | Duration, as printed 14m 02s |
| SAGE (Vals AI) | 53.6% | Not published | Latency, as printed 1m 03s |
| LegalBench (Vals AI) | 88.3% | Not published | Latency, as printed 13 s |
| MedCode (Vals AI) | 58.8% | Not published | Latency, as printed 1m 20s |
| MedScribe (Vals AI) | 87.4% | Not published | Latency, as printed 1m 27s |
| APEX-Agents | 82.4% | Not published | Mean score 87.4% |
| LMArena Text | 1532 Elo | Not published | Votes 4,904 · Interval (±) ±8.8 |
| LMArena Text: coding | 1547 Elo | Not published | Votes 1,231 · Interval (±) ±17.1 |
| LMArena Text: hard prompts | 1546 Elo | Not published | Votes 3,129 · Interval (±) ±10.9 |
| LMArena Text: math | 1536 Elo | Not published | Votes 273 · Interval (±) ±34.3 |
| LMArena Text: instruction following | 1536 Elo | Not published | Votes 1,750 · Interval (±) ±14.6 |
| LMArena Text: creative writing | 1527 Elo | Not published | Votes 1,057 · Interval (±) ±19.4 |
| LMArena Text: English | 1538 Elo | Not published | Votes 1,902 · Interval (±) ±13.8 |
| LMArena Text: non-English | 1522 Elo | Not published | Votes 3,002 · Interval (±) ±11.1 |
| LMArena Text: Chinese | 1608 Elo | Not published | Votes 363 · Interval (±) ±31.4 |
| LMArena Text: Spanish | 1504 Elo | Not published | Votes 225 · Interval (±) ±38.2 |
| LMArena Text: Russian | 1534 Elo | Not published | Votes 671 · Interval (±) ±22.7 |
| LMArena Text: multi-turn | 1556 Elo | Not published | Votes 680 · Interval (±) ±23.3 |
| LMArena Text: longer queries | 1546 Elo | Not published | Votes 2,064 · Interval (±) ±13.6 |
| LMArena Text: expert | 1551 Elo | Not published | Votes 510 · Interval (±) ±25.9 |
| LMArena Text: hard prompts in English | 1548 Elo | Not published | Votes 1,208 · Interval (±) ±17.3 |
| LMArena Text: software and IT | 1556 Elo | Not published | Votes 1,822 · Interval (±) ±14.1 |
| LMArena Text: mathematical work | 1556 Elo | Not published | Votes 328 · Interval (±) ±31.2 |
| LMArena Text: science | 1548 Elo | Not published | Votes 819 · Interval (±) ±21.1 |
| LMArena Text: medicine and healthcare | 1554 Elo | Not published | Votes 334 · Interval (±) ±32.8 |
| LMArena Text: legal and government | 1549 Elo | Not published | Votes 402 · Interval (±) ±30.1 |
| LMArena Text: business and finance | 1525 Elo | Not published | Votes 891 · Interval (±) ±20.3 |
| LMArena Text: writing and language | 1529 Elo | Not published | Votes 1,365 · Interval (±) ±16.8 |
| LMArena Text: entertainment and media | 1526 Elo | Not published | Votes 1,233 · Interval (±) ±18.0 |
| LMArena WebDev: React | 1692 Elo | Not published | Votes 1,043 · Interval (±) ±20.5 |
| LMArena WebDev: HTML | 1643 Elo | Not published | Votes 185 · Interval (±) ±48.3 |