Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Google · Proprietary · rank 10 · bench-align-v5
Gemini 4 Argon is built for coding, agent work and reasoning. It ranks in the top quarter for reasoning (top 4% of 216 scored models) and agent work (top 12% of 127 scored models), by BenchLM's published scores.
Every published price component, including cache reads and writes.
Lifecycle events the source associates with this model, plus what this profile does not claim.
Gemini 4 Argon by Google scores 79.1 overall, costs $2.00 per 1M input tokens and $10.00 per 1M output tokens, and carries a context window of 1M tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Each axis shows the ranking source’s score and, beside it, BenchLM’s separate verified score for that axis. The two are published side by side and are not interchangeable.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: Observed
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | 47.13 s | 2026-11-21 | 2026-11-21 | |
| Throughput | 65 tok/s | 2026-11-21 | 2026-11-21 |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
| $5.00 |
| Batch cache read / 1M tokens (OpenRouter batch) | $0.05 |
| Batch cache write / 1M tokens (OpenRouter batch) | $0.19 |
Batch API rate (asynchronous jobs), from OpenRouter's batch listing.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Plans from the model's maker, and the API providers that list this exact model today. Links leave ModelScale; each provider sets its own prices and limits.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| LABBench2 LABBench2: An Improved Benchmark for AI Systems Performing Biology Research | 88.8 | #1 | Nearly 1,900 biology-research tasks | Real-world biology research | LABBench2: An Improved Benchmark for AI Systems Performing Biology Research |
| Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index | 50.4 | #18 | Cross-benchmark intelligence index | Display-only external reference | Artificial Analysis |
| AA-HLE Artificial Analysis Humanity's Last Exam | 57.1 | #4 | Expert-level questions | Frontier expert reasoning | Artificial Analysis Humanity's Last Exam Benchmark Leaderboard |
| AA-Omniscience Index Artificial Analysis Omniscience Index | 46.1 | #8 | Knowledge questions | Broad factual knowledge | AA-Omniscience: Knowledge and Hallucination Benchmark |
| AA-Omniscience Accuracy Artificial Analysis Omniscience Accuracy | 48.3 | #33 | Knowledge questions | Broad knowledge | Artificial Analysis model benchmarks |
| AA-Omniscience Hallucination Rate Artificial Analysis Omniscience Hallucination Rate | 4.3 | #218 | Knowledge questions | Factuality | Artificial Analysis model benchmarks |
| Morphology-to-molecule matching Axiom Bio morphology-to-molecule matching | 25.2 | #3 | Four multi-channel cell-imaging perturbation-matching tasks | Biomedical image and molecular reasoning | Claude Opus 5.5 System Card |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Vibe Code Bench Vibe Code Bench v1.1 | 91.90 | #1 | End-to-end web application builds | End-to-end software delivery | Vibe Code Bench: Evaluating AI Models on End-to-End Web Application Development |
| FrontierSWE v2 FrontierSWE v2 | 55.0 | #7 | 34 ultra-long-horizon engineering and research tasks | Ultra-long-horizon frontier software engineering | FrontierSWE v2 |
| Bug Hunt Bench Bug Hunt Bench | 40.7 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3) FrontierMath v2 Tiers 1-3 | 85.965 | #15 | 285 private advanced mathematics problems | From olympiad-plus to early research mathematics | FrontierMath v2 benchmark hub |
| FrontierMath v2 (Tier 4) FrontierMath v2 Tier 4 | 78.049 | #14 | 41 private extreme-difficulty mathematics problems | Research-level mathematics requiring hours or days of expert work | FrontierMath Tier 4 v2 leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| ARC-AGI-1 ARC-AGI-1 Semi-Private Evaluation | 98.50 | #1 | Semi-private ARC-AGI-1 evaluation set | Abstract visual reasoning | ARC Prize leaderboard |
| Graphwalks BFS 128K Graphwalks BFS 0K-128K | 99.7 | #1 | Graph traversal tasks | Algorithmic long-context reasoning | Introducing GPT-5.4 mini and nano |
| ARC-AGI-2 Abstraction and Reasoning Corpus for AGI v2 | 91.25 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 4.0 Terminal-Bench 4.0 | 57.40 | #8 | 66 professional computer-work tasks | Frontier autonomous knowledge work | Terminal-Bench 4.0 |
| AA Briefcase Artificial Analysis Briefcase | 1486 | #29 | Professional knowledge-work tasks | Professional work | Artificial Analysis Briefcase Benchmark Leaderboard |
| AA AutomationBench Artificial Analysis AutomationBench | 77.5 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Chartography (no tools) Chartography without tools | 71.6 | #3 | 100 specialized chart types | Professional chart reasoning | Chartography |
| Design Arena Website Design Arena Website Elo | 1313 | #9 | Website generation comparisons | Design and website generation | OpenRouter Grok 4.3 benchmarks |
| LVBench LVBench | 91.7 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| SEC-Bench Pro SEC-Bench Pro | 83.4 | #6 | Security engineering tasks | Advanced cybersecurity | Introducing GPT-5.6 |
| Chess Puzzles Chess Puzzles (Epoch AI) | 59.0 | #11 | 100 programmatically generated chess positions with a single best move | Novel chess positions, single best move | Chess Puzzles (Epoch AI) |
| Mystery Game Puzzles Mystery Game Puzzles (Epoch AI) | 64.0 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Gray Swan IPI (15 attempts) Gray Swan Indirect Prompt Injection, attack success at 15 attempts | 0.7 | #13 | Indirect prompt-injection attacks | Adversarial prompt injection | Gemini 4 Argon evaluation methodology |
A second, independent score source. Epoch’s own figures in Epoch’s units, one line per model version Epoch evaluated (effort setting as Epoch labels it). They are not part of this site’s capability scores or rank.
| Benchmark | Variant | Score | Epoch date | Evidence |
|---|---|---|---|---|
| GPQA Diamond Accuracy on graduate-level science multiple-choice questions (best score across scorers). Unit: fraction (0–1); higher is better. | high | 0.964 ± 0.0119 | 2026-10-13 | |
| FrontierMath Tiers 1-3 (v2) Share of private research-level math problems solved, tiers 1-3. Unit: fraction (0–1); higher is better. | high | 0.8596 ± 0.0206 | 2026-10-13 | |
| FrontierMath Tier 4 (v2) Share of private research-level math problems solved, tier 4. Unit: fraction (0–1); higher is better. | high | 0.7805 ± 0.0654 | 2026-10-13 | |
| SimpleQA Verified Short factual questions answered correctly. Unit: fraction (0–1); higher is better. | high | 0.76 ± 0.0135 | 2026-10-13 |
Source: Epoch AI, CC BY 4.0 (licence), benchmark data read once a day.
| #12 |
| 105 planted bugs across two production TypeScript repositories |
| Blind production-repository bug finding and repair |
| Bug Hunt Bench method, caveats, and definitions |
| AA-SciCode Artificial Analysis SciCode | 61.8 | #5 | Scientific coding subproblems | Scientific programming | Artificial Analysis SciCode Benchmark Leaderboard |
| DeepSWE DeepSWE | 77.9 | #3 | 113 software engineering tasks across 91 repositories and 5 languages | Long-horizon software engineering | DeepSWE benchmark blog |
| PostTrainBench v1.1 PostTrainBench v1.1 | 45.3 | #2 | Post-training Qwen3 1.7B and 4B, SmolLM3 3B, and Gemma3 4B | Frontier agent evaluation | PostTrainBench v1.1 leaderboard and methodology |
| FrontierCode 2.0 FrontierCode 2.0 | 27.6 | #18 | 132 private software-engineering tasks | Frontier coding-agent quality | FrontierCode 2.0 |
| ScarfBench ScarfBench: Cross-Framework Application Migration in Enterprise Java | 75.2 | #1 | 204 framework migration tasks across 34 enterprise Java applications | Whole-application framework migration | ScarfBench: A Benchmark for Cross-Framework Application Migration in Enterprise Java |
| PostTrainBench PostTrainBench v1.2 | 40.4 | #4 | Post-training four base models on six benchmarks | Autonomous ML engineering | PostTrainBench |
| #7 |
| Visual pattern completion and abstract reasoning |
| Expert-level — hardest public reasoning benchmark |
| ARC-AGI-2: A Harder General Intelligence Benchmark |
| ARC-AGI-3 Abstraction and Reasoning Corpus for AGI v3 | 46.265668 | #7 | Interactive game-like tasks with hidden rules | Frontier agentic reasoning | ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence |
| AA-LCR Artificial Analysis Long Context Reasoning | 79.7 | #48 | Long-context reasoning tasks | Long-context reasoning | Artificial Analysis model benchmarks |
| CritPt Critical Physics Tasks | 27.1 | #17 | Research-level physics questions | Research-level physics reasoning | CritPt Benchmark Leaderboard |
| GraphWalks BFS 256K–1M GraphWalks BFS 256K–1M | 84.2 | #2 | Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens. | Frontier agent evaluation | Gemini 4 Argon evaluation methodology |
| STEMBench STEMBench120: six-domain scientific research progress | 2.44 | #8 | 120 research tasks, 20 in each of six STEM domains | Open research targets; no run has reached a full-target success | STEMBench: six-domain model evaluation |
| ARC-AGI-3 Provider Adapter ARC-AGI-3 Semi-Private: Provider Adapter harness | 97.30 | #4 | — | — | ARC Prize verified ARC-AGI-3 results |
| #3 |
| Business-process automation tasks |
| Agentic automation |
| Artificial Analysis AutomationBench Benchmark Leaderboard |
| GDPval-AA GDPval-AA | 1624 | #25 | Agentic real-world work tasks | Professional agentic workflows | DeepSeek-V4 Technical Report |
| GDPval-AA GDPval-AA normalized | 56.2 | #14 | Economically valuable tasks | Professional agentic workflows | Artificial Analysis model benchmarks |
| AA Terminal-Bench 4.0 Artificial Analysis Terminal-Bench v4.0 | 57.1 | #10 | Terminal-based agent tasks | Agentic software engineering | Artificial Analysis Terminal-Bench v4.0 Benchmark Leaderboard |
| GDP.pdf Artificial Analysis GDP.pdf | 21.8 | #27 | Professional document-production tasks | Professional knowledge work | GDP.pdf Benchmark Leaderboard |
| AA-AnalystAgent Artificial Analysis AnalystAgent | 48.8 | #14 | Spreadsheet and document analysis questions | Business and data analysis | AA-AnalystAgent Benchmark Leaderboard |
| OSWorld 2.0 OSWorld 2.0 | 69.2 | #4 | 108 long-horizon computer-use workflows | Long-horizon professional workflows | OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks |
| ExploitGym ExploitGym | 39.5 | #4 | 898 exploitation tasks | Advanced cybersecurity exploitation | ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? |
| AutomationBench AutomationBench | 51.3 | #13 | 600 public automation tasks | Long-horizon automation | Kimi K3: Open Frontier Intelligence |
| Agents' Last Exam Agents' Last Exam | 39.5 | #9 | Agent tasks | Advanced agentic work | DeepSeek V4 Flash 0731 update |
| Finance Agent v2 Finance Agent v2 | 65.4 | #1 | Financial analyst task categories | Professional expert-task agent workflow | Finance Agent v2 |
| Terminal-Bench 2.1 (Vals) Terminal-Bench 2.1, Vals AI run | 57.7 | #41 | Difficult terminal tasks | Frontier agentic | Vals AI Terminal-Bench 2.1, Vals AI run leaderboard |
| Terminal-Bench-Science 0.1 (6x verifier timeout) Terminal-Bench-Science 0.1 with sixfold verifier timeout | 57.6 | #1 | 70 expert-curated scientific research workflows | Frontier scientific research workflows | Gemini 4 Argon evaluation methodology |
| CWE-bench v1 CWE-bench v1 | 68.0 | #2 | Audit and patch held-out C/C++, Go, Java, TypeScript/JavaScript, Python, and Rust repositories | Frontier agent evaluation | CWE-bench v1 leaderboard and methodology |
| MazeBench MazeBench: visual spatial reasoning in a 3D open world | 14 | #9 | 100 hidden gems across 256 rooms in one open world | Long-horizon spatial planning, with single puzzles that can take over one hundred moves | MazeBench Results |
| RoboDojo Sim RoboDojo Sim | 34.2 | #3 | Simulated robot-control tasks (count not published) | Embodied agent control | Introducing Claude Fable 5.2 and Claude Mythos 5.2 |
| CWE-bench CWE-bench v2 | 65.0 | #5 | Audit and patch C/C++, Go, Java, TypeScript/JavaScript, Python, and Rust repositories; 100 public and 150 held-out tasks | Frontier agent evaluation | CWE-bench v2 leaderboard and methodology |
| Terminal-Bench 4.1 Terminal-Bench 4.1 | 55.50 | #12 | 66 professional computer-work tasks | Frontier autonomous knowledge work | Terminal-Bench 4.1 |
| OSWorld 2.1 Offline OSWorld 2.1 offline subset (82 tasks) | 74.7 | #7 | 82 offline computer-use workflows | Long-horizon professional workflows | OSWorld 2.1 release (xlang-ai/OSWorld-V2, tag osworld-v2.1) |
| MKA Multimodal Knowledge Agent Benchmark | 51.3 | #8 | 200 multimodal professional tasks across 10 industry verticals | Expert multimodal knowledge work | MKA Benchmark Leaderboard |
| #1 |
| Long-form video question answering |
| Extended temporal reasoning |
| Qwen3.8-Max: A New Bar for Coding and Cowork |
| Blueprint-Bench 2 Blueprint-Bench 2 | 54.4 | #2 | Spatial reasoning from blueprints | Agentic spatial reasoning | Gemini 3.5 Flash launch screenshots |
| ExtractBench Pro ExtractBench Pro: long-tail document extraction challenges (LlamaIndex) | 30 | #5 | 100 synthetic documents, 2,622 pages, 86 document types, six challenge groups | The best system completes 58 of 100 documents perfectly; several complete none | ExtractBench Pro leaderboard and challenge descriptions |
| ParseBench-2 ParseBench-2: severity-aware document parsing (LlamaIndex leaderboard) | 85.17 | #2 | 4,317 page-level test cases from 2,645 documents in 13 datasets | Charts and visual grounding separate the field; several systems that lead on text and tables score near zero on charts | ParseBench-2: Severity-Aware Evaluation of Document Parsing |
| #10 |
| 100 single-best-move positions from an undisclosed game |
| Unfamiliar game positions, single best move |
| Mystery Game Puzzles (Epoch AI) |
| Furniture Assembly Furniture Assembly Benchmark (Epoch AI) | 72.5 | #9 | 60 photos of three IKEA builds, some with deliberate assembly mistakes | Spatial reasoning from technical diagrams to real photos | Furniture Assembly Benchmark (Epoch AI) |
| EBR-bench Earthborne Rangers (EBR-bench) | 64.8 | #8 | Ten playthroughs of an Earthborne Rangers campaign segment per sample (eight learning, two scored) | Learning an unfamiliar game over multi-hour playthroughs | Earthborne Rangers (EBR-bench), Epoch AI |
| MirrorCode MirrorCode (Epoch AI and METR) | 58.9 | #14 | 30 reimplementation tasks (15 programs in two languages), three attempts each | Days-long software engineering without source code | MirrorCode (Epoch AI) |
| FACTS Multimodal FACTS Multimodal | 54.2 | #5 | About 1,500 image questions: 711 public and 811 held out | Image-grounded factual answering | BenchLM |
| DAYJOB DAYJOB | 15.7 | #5 | 130 long-horizon tasks: 80 finance and 50 healthcare | Long-horizon professional work | BenchLM |
| DAYJOB Healthcare DAYJOB Healthcare | 11.6 | #4 | 50 long-horizon healthcare tasks across five subfields | Long-horizon clinical and healthcare operations work | BenchLM |
| Code2Video Code2Video | 1539 | #7 | Creative briefs rendered as motion-graphics video | Code-driven motion design | BenchLM |
| Landing Page Arena Landing Page Arena | 1385 | #13 | 12 client briefs for small-business landing pages, rated by human votes | Marketing page design and copy | BenchLM |
| Chart Arena Chart Arena | 1528 | #6 | 13 chart briefs with small datasets, rated by human votes | Data visualisation | BenchLM |
| Joke Arena Joke Arena | 1494 | #10 | 100 joke prompts in 10 categories, rated by human votes | Humour writing | BenchLM |
| Pelican Arena Pelican Arena | 1680 | #5 | One prompt: an animated SVG of a pelican riding a bicycle, rated by human votes | Vector drawing and animation in code | BenchLM |
| Landmark Arena Landmark Arena | 1610 | #4 | 20 famous landmarks built as three.js web pages, rated by human votes | 3D scenes in browser code | BenchLM |
| ASCII Arena ASCII Arena | 1981 | #2 | ASCII drawings of subjects from Google's Quick, Draw! dataset, rated by human votes | Text-character drawing | BenchLM |
| Game Arena Go Game Arena Go | 1686 | #4 | Head-to-head Go games on a 13 x 13 board | Strategic planning and rule-following | BenchLM |
| NYT Connections NYT Connections LLM Benchmark (extended version) | 98.6 | #2 | 940 NYT Connections puzzles, each with up to four added trick words | Wordplay, word knowledge and decoys that fit almost-but-not-quite | NYT Connections LLM Benchmark (Lech Mazur, GitHub) |
| StealthMark StealthMark Intelligence Benchmarks (overall score) | 67.4 | #5 | 8 original tasks: Newton's Mirror, Ground Truth, Latch, Foresight, Mona Lisa, Selfie, Beach and Raccoon | Chained physical reasoning, planning without tools and precise visual construction | StealthMark: The AI Benchmark Explained (Stealth Models) |
| FrontierPhysics End-to-End Physics Research Benchmark for AI Agents | 8.3 | #8 | 56 research projects from 42 physicists across nine fields, three trials each (168 rollouts per setup) | Open-ended research projects that took their authors about 270 days on average | FrontierPhysics: Evaluating Agents for End-to-End Physics Research (arXiv, November 4, 2026) |
| HoneyBench HoneyBench v0.2 (Goodhart Labs) | 63.6 | #14 | 20 private honeypot tasks across 7 domains with exploitable evaluators | Agentic alignment and specification gaming | HoneyBench v0.2 |