# Google Gemini 4 Argon As of October 22, 2026, Google Gemini 4 Argon ranks #3 of 123 models overall with 84.0 ±3.8 on the LLMBenchmark Index. Gemini 4 Argon ranks #1 of 122 in reasoning, its strongest category. Google Gemini 4 Argon costs $2.00 per million input tokens and $10.00 per million output tokens. Gemini 4 Argon was released on October 13, 2026. Page: https://llmbenchmark.ai/models/gemini-4-argon Data from October 22, 2026. Every figure below names the board it was read from and the day. - Maker: Google - Released: October 13, 2026 - Price per million input tokens: $2.00 - Price per million output tokens: $10.00 - Context window: 1.05M tokens - Overall rank: 3 of 123, by the LLMBenchmark Index - LLMBenchmark Index: 84.0 ±3.8 (90% margin, from 7 of 7 inputs) - Artificial Analysis Intelligence Index: 52.6 - Vibe Coding Index: 81.0 (from 5 of 6 coding benchmarks) Benchmark results (94 of 195): - LiveBench coding: 89.7% (source: LiveBench, read October 22, 2026) - LMArena WebDev: 1678 Elo (source: LMArena WebDev, read October 22, 2026) - Vals Index: 69.2% (source: Vals Index, read October 22, 2026) - AA Intelligence Index: 52.6 (source: Artificial Analysis (via OpenRouter), read October 22, 2026) - Design Arena: websites: 1315 Elo (source: Design Arena, read October 22, 2026) - Design Arena: UI components: 1333 Elo (source: Design Arena, read October 22, 2026) - Design Arena: game dev: 1377 Elo (source: Design Arena, read October 22, 2026) - Design Arena: data viz: 1332 Elo (source: Design Arena, read October 22, 2026) - Design Arena: 3D: 1388 Elo (source: Design Arena, read October 22, 2026) - Design Arena: all code categories: 1329 Elo (source: Design Arena, read October 22, 2026) - Design Arena: spreadsheets: 1157 Elo (source: Design Arena, read October 22, 2026) - LiveBench reasoning: 95.1% (source: LiveBench, read October 22, 2026) - LiveBench data analysis: 85.2% (source: LiveBench, read October 22, 2026) - LiveBench research: 83.1% (source: LiveBench, read October 22, 2026) - LiveBench browser: 92.0% (source: LiveBench, read October 22, 2026) - LiveBench long agentic: 69.9% (source: LiveBench, read October 22, 2026) - LiveBench design: 73.7% (source: LiveBench, read October 22, 2026) - Terminal-Bench 4 (Vals AI): 57.6% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench (Vals AI): 91.9% (source: Vals AI boards, read October 22, 2026) - Code migration (Vals AI): 68.2% (source: Vals AI boards, read October 22, 2026) - ProgramBench (Vals AI): 2.5% (source: Vals AI boards, read October 22, 2026) - SREBench (Vals AI): 44.3% (source: Vals AI boards, read October 22, 2026) - IOI (Vals AI): 100.0% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench 2.1 (Vals AI): 57.7% (source: Vals AI boards, read October 22, 2026) - SWE-bench Multimodal (Vals AI): 56.0% (source: Vals AI boards, read October 22, 2026) - Vibe Code Bench 1-100 (Vals AI): 21.2% (source: Vals AI boards, read October 22, 2026) - CyberBench (Vals AI): 77.9% (source: Vals AI boards, read October 22, 2026) - Finance Agent (Vals AI): 65.4% (source: Vals AI boards, read October 22, 2026) - Tax Agent Bench (Vals AI): 44.6% (source: Vals AI boards, read October 22, 2026) - Legal Research Bench (Vals AI): 54.8% (source: Vals AI boards, read October 22, 2026) - Harvey's Legal Agent Benchmark (Vals AI): 27.1% (source: Vals AI boards, read October 22, 2026) - Excel Modeling Benchmark (Vals AI): 75.2% (source: Vals AI boards, read October 22, 2026) - BioMysteryBench (Vals AI): 76.3% (source: Vals AI boards, read October 22, 2026) - Terminal-Bench-Science (Vals AI): 44.3% (source: Vals AI boards, read October 22, 2026) - Vals Science Index: 55.4% (source: Vals AI boards, read October 22, 2026) - Time Horizon Index (Vals AI): 52.5% (source: Vals AI boards, read October 22, 2026) - CUA-bench (Vals AI): 4.8% (source: Vals AI boards, read October 22, 2026) - Robo-bench (Vals AI): 0.8% (source: Vals AI boards, read October 22, 2026) - Public Benefits Bench (Vals AI): 69.8% (source: Vals AI boards, read October 22, 2026) - MysteryMechanism (Vals AI): 45.5% (source: Vals AI boards, read October 22, 2026) - SAGE (Vals AI): 53.6% (source: Vals AI boards, read October 22, 2026) - LegalBench (Vals AI): 88.3% (source: Vals AI boards, read October 22, 2026) - MedCode (Vals AI): 58.8% (source: Vals AI boards, read October 22, 2026) - MedScribe (Vals AI): 87.4% (source: Vals AI boards, read October 22, 2026) - Humanity's Last Exam (Artificial Analysis run): 57.1% (source: Artificial Analysis, read October 22, 2026) - SciCode: 61.8% (source: Artificial Analysis, read October 22, 2026) - Long context reasoning (AA-LCR): 79.7% (source: Artificial Analysis, read October 22, 2026) - FrontierCode: 47.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierSWE: 55.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - MirrorCode: 58.9% (source: Epoch AI benchmarking hub, read October 22, 2026) - APEX-Agents: 82.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - GPQA Diamond (Epoch AI run): 96.4% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI: 98.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - ARC-AGI-2: 91.3% (source: Epoch AI benchmarking hub, read October 22, 2026) - Chess puzzles: 59.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - Mystery game puzzles: 64.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - EBR-bench: 64.8% (source: Epoch AI benchmarking hub, read October 22, 2026) - OTIS Mock AIME: 100.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tiers 1 to 3: 86.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - FrontierMath, tier 4: 78.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - ProofBench: 99.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - SimpleQA Verified: 76.0% (source: Epoch AI benchmarking hub, read October 22, 2026) - DTBench: 97.6% (source: Epoch AI benchmarking hub, read October 22, 2026) - LMCA: 63.7% (source: Epoch AI benchmarking hub, read October 22, 2026) - Furniture assembly: 72.5% (source: Epoch AI benchmarking hub, read October 22, 2026) - ALE-Bench: 2100.1 (source: ALE-Bench (Sakana AI), read October 22, 2026) - Vending-Bench 2: $13,718 (source: Epoch AI benchmarking hub, read October 22, 2026) - Epoch Capabilities Index: 166.4 (source: Epoch AI benchmarking hub, read October 22, 2026) - LMArena Text: 1532 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: coding: 1546 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts: 1546 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: math: 1536 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: instruction following: 1534 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: creative writing: 1527 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: English: 1538 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: non-English: 1522 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Chinese: 1611 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Spanish: 1503 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: Russian: 1534 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: multi-turn: 1556 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: longer queries: 1547 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: expert: 1550 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: hard prompts in English: 1548 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: software and IT: 1556 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: mathematical work: 1557 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: science: 1548 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: medicine and healthcare: 1555 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: legal and government: 1548 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: business and finance: 1525 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: writing and language: 1529 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena Text: entertainment and media: 1527 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: React: 1692 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - LMArena WebDev: HTML: 1643 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026) - SciCode: 12.3% (source: Kaggle Benchmarks, read October 22, 2026)