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 60 · bench-align-v5
Gemini 3 Pro is built for coding. It ranks in the top quarter for knowledge (top 7% of 284 scored models) and coding (top 10% of 210 scored models), by BenchLM's published scores.
Gemini 3 Pro by Google scores 59.8 overall, costs $2.00 per 1M input tokens and $12.00 per 1M output tokens, and carries a context window of 2M tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Every published price component, including cache reads and writes.
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.
Lifecycle events the source associates with this model, plus what this profile does not claim.
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 | 32.65 s | 2026-11-21 | 2026-11-21 | |
| Throughput | 109 tok/s | 2026-11-21 | 2026-11-21 |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
Plans from the model's maker, and the API providers that list this exact model today. Links leave ModelChoice; each provider sets its own prices and limits.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA MMLU-Pro Artificial Analysis MMLU-Pro | 89.8 | #1 | Professional multi-subject questions | Professional knowledge and reasoning | Artificial Analysis MMLU-Pro Benchmark Leaderboard |
| Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index | 28.0 | #69 | Cross-benchmark intelligence index | Display-only external reference | Artificial Analysis |
| AA-GPQA Diamond Artificial Analysis GPQA Diamond | 90.8 | #33 | Graduate-level science questions | Graduate-level science reasoning | Artificial Analysis GPQA Diamond Benchmark Leaderboard |
| AA-HLE Artificial Analysis Humanity's Last Exam | 39.7 | #51 | Expert-level questions | Frontier expert reasoning | Artificial Analysis Humanity's Last Exam Benchmark Leaderboard |
| AA-Omniscience Index Artificial Analysis Omniscience Index | 15.3 | #43 | Knowledge questions | Broad factual knowledge | AA-Omniscience: Knowledge and Hallucination Benchmark |
| AA-Omniscience Accuracy Artificial Analysis Omniscience Accuracy | 55.8 | #14 | Knowledge questions | Broad knowledge | Artificial Analysis model benchmarks |
| AA-Omniscience Hallucination Rate Artificial Analysis Omniscience Hallucination Rate | 91.5 | #28 | Knowledge questions | Factuality | Artificial Analysis model benchmarks |
| GPQA Diamond (Vals) GPQA Diamond, Vals AI run | 91.7 | #19 | Graduate-level science questions | Expert reasoning | Vals AI GPQA Diamond, Vals AI run leaderboard |
| MMLU-Pro (Vals) MMLU-Pro, Vals AI run | 90.1 | #6 | Academic multiple-choice questions | Broad academic knowledge | Vals AI MMLU-Pro, Vals AI run leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA LiveCodeBench Artificial Analysis LiveCodeBench | 91.7 | #1 | Contamination-resistant coding tasks | Competitive programming | Artificial Analysis LiveCodeBench Benchmark Leaderboard |
| Vibe Code Bench Vibe Code Bench v1.1 | 14.30 | #34 | End-to-end web application builds | End-to-end software delivery | Vibe Code Bench: Evaluating AI Models on End-to-End Web Application Development |
| LiveCodeBench (Vals) LiveCodeBench, Vals AI run | 86.4 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA AIME 2025 Artificial Analysis AIME 2025 | 95.7 | #2 | 30 AIME 2025 problems | Olympiad mathematics | Artificial Analysis AIME 2025 Benchmark Leaderboard |
| FrontierMath v2 (Tiers 1-3) FrontierMath v2 Tiers 1-3 | 37.600 | #55 | 285 private advanced mathematics problems | From olympiad-plus to early research mathematics | FrontierMath v2 benchmark hub |
| FrontierMath v2 (Tier 4) FrontierMath v2 Tier 4 | 18.750 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| ARC-AGI-1 ARC-AGI-1 Semi-Private Evaluation | 75.00 | #39 | Semi-private ARC-AGI-1 evaluation set | Abstract visual reasoning | ARC Prize leaderboard |
| ARC-AGI-2 Abstraction and Reasoning Corpus for AGI v2 | 31.1 | #46 | Visual pattern completion and abstract reasoning | Expert-level — hardest public reasoning benchmark | ARC-AGI-2: A Harder General Intelligence Benchmark |
| AA-LCR Artificial Analysis Long Context Reasoning | 76.0 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA-IFBench Artificial Analysis IFBench | 70.4 | #45 | Verifiable instruction constraints | Instruction precision | Artificial Analysis IFBench Benchmark Leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA Global-MMLU-Lite Artificial Analysis Global-MMLU-Lite | 92.2 | #3 | Multilingual knowledge questions | Multilingual professional knowledge | Artificial Analysis Global-MMLU-Lite Benchmark Leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Gert Labs Gert Labs Composite Game Benchmark | 63.23 | #7 | Novel game environments | Agentic coding and decision-making | Gert Labs rankings |
| JobBench JobBench | 11.4 | #35 | 130 tasks across 35 occupations | Professional multi-source workflows | JobBench: Aligning Agent Work With Human Will |
| τ²-bench results τ²-Bench Tool-Agent-User Evaluation | 87.1 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| MMMU Massive Multi-discipline Multimodal Understanding | 87.2 | #1 | Multimodal academic reasoning | Frontier multimodal | MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI |
| MMMU-Pro Massive Multi-discipline Multimodal Understanding Pro | 81 | #12 | Multimodal academic reasoning | Frontier multimodal | MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark |
| AA-MMMU-Pro Artificial Analysis MMMU-Pro | 80.2 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FailureSensorIQ FailureSensorIQ | 69.1 | #2 | Multiple-choice questions on sensors and failure modes for 10 industrial assets | Industrial maintenance and reliability knowledge | BenchLM |
| BinaryAudit BinaryAudit (backdoor detection in compiled binaries) | 43.9 | #4 | 19 backdoor and timebomb detection tasks on stripped binaries, three runs each (57 runs); 6 tooling and 8 clean-binary tasks reported alongside | Reverse engineering without source code or symbols | We hid backdoors in ~40 MB binaries and asked AI + Ghidra to find them (Quesma) |
| OTelBench OTelBench (OpenTelemetry instrumentation tasks) |
People's votes between two anonymous answers on arena.ai, one board per task. LMArena's own ratings and ranks, one line per model version it rates; not part of this site's capability scores or rank.
| Board | Variant | Rating | Votes | Rank | Published | Evidence |
|---|---|---|---|---|---|---|
| Vision Arena People's votes between two answers about the same image, style-controlled. | — | 1,289 ± 7 | 13,482 | 15 | 2026-11-19 |
Source: LMArena, CC BY 4.0 (licence), leaderboard data read once a day.
| #22 |
| Competitive programming problems (easy, medium, hard) |
| Frontier coding |
| Vals AI LiveCodeBench, Vals AI run leaderboard |
| SWE-bench (Vals) SWE-bench, Vals AI run | 76.4 | #36 | Real repository issues by human time bucket | Frontier coding agents | Vals AI SWE-bench, Vals AI run leaderboard |
| PostTrainBench v1 PostTrainBench v1 (archived) | 18.1 | #10 | Post-training four base models on seven benchmarks | Autonomous ML engineering | PostTrainBench v1 leaderboard (archived) |
| #45 |
| 41 private extreme-difficulty mathematics problems |
| Research-level mathematics requiring hours or days of expert work |
| FrontierMath Tier 4 v2 leaderboard |
| #93 |
| Long-context reasoning tasks |
| Long-context reasoning |
| Artificial Analysis model benchmarks |
| CritPt Critical Physics Tasks | 9.1 | #73 | Research-level physics questions | Research-level physics reasoning | CritPt Benchmark Leaderboard |
| Airline, retail, and telecom customer-service task sets |
| Dual-control customer-service workflows |
| τ²-Bench: Evaluating Conversational Agents in a Dual-Control Environment |
| #23 |
| Multimodal academic reasoning |
| Frontier multimodal |
| Artificial Analysis MMMU-Pro Benchmark Leaderboard |
| OmniDocBench 1.5 OmniDocBench 1.5 | 88.5 | #8 | Document understanding tasks | Grounded document reasoning | Introducing GPT-5.4 mini and nano |
| RealWorldQA RealWorldQA | 83.3 | #9 | Real-world visual question answering | General visual reasoning | Qwen3.6 launch benchmarks |
| Video-MME (with subtitle) Video-MME with subtitle | 88.4 | #2 | Video understanding | Multimodal video reasoning | Qwen3.6 launch benchmarks |
| Video-MME (w/o subtitle) Video-MME without subtitle | 87.7 | #1 | Video understanding | Multimodal video reasoning | Qwen3.6 launch benchmarks |
| MathVision MathVision | 86.6 | #15 | Visually grounded math problems | Advanced multimodal mathematics | Qwen3.6 launch benchmarks |
| We-Math We-Math | 86.9 | #2 | Visually grounded math problems | Advanced multimodal mathematics | Qwen3.6 launch benchmarks |
| DynaMath DynaMath | 85.1 | #4 | Dynamic visual math problems | Advanced multimodal mathematics | Qwen3.6 launch benchmarks |
| MMLongBench-Doc MMLongBench-Doc | 60.5 | #4 | Long document understanding | Long-context document reasoning | Qwen3.6 launch benchmarks |
| CC-OCR CC-OCR | 79.0 | #7 | Optical character recognition | Document reading | Qwen3.6 launch benchmarks |
| AI2D_TEST AI2D test split | 94.1 | #1 | Diagram understanding | Structured visual reasoning | Qwen3.6 launch benchmarks |
| CountBench CountBench | 97.3 | #2 | Visual counting tasks | Fine-grained visual perception | Qwen3.6 launch benchmarks |
| RefCOCO (avg) RefCOCO average | 84.1 | #9 | Referring-expression grounding | Fine-grained visual grounding | RefCOCO referring expression datasets |
| ODINW13 ODINW13 | 46.3 | #4 | Out-of-distribution object understanding | Robust visual grounding | Qwen3.6 launch benchmarks |
| ERQA ERQA | 70.5 | #7 | Evidence-based visual QA | Grounded multimodal reasoning | Qwen3.6 launch benchmarks |
| VideoMMMU VideoMMMU | 87.6 | #2 | Video-grounded expert reasoning | Frontier multimodal video reasoning | Qwen3.6 launch benchmarks |
| MLVU (M-Avg) MLVU mean average | 83.0 | #8 | General video understanding | Broad multimodal video reasoning | Qwen3.6 launch benchmarks |
| ScreenSpot Pro ScreenSpot Pro | 72.7 | #12 | 1,581 grounding instructions | Professional GUI grounding | ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use |
| SimpleVQA SimpleVQA | 73.2 | #5 | Visual QA tasks | General visual understanding | GLM-5V-Turbo |
| V* V* | 88.0 | #9 | Frontier multimodal reasoning tasks | Frontier multimodal | GLM-5V-Turbo |
| CharXiv CharXiv Reasoning | 81.4 | #31 | Scientific chart reasoning | Scientific visualization reasoning | CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs |
| #5 |
| 23 OpenTelemetry instrumentation tasks in 11 languages, three attempts each (69 runs) |
| Distributed tracing and context propagation across services |
| Benchmarking OpenTelemetry: can AI trace your failed login? (Quesma) |
| NYT Connections NYT Connections LLM Benchmark (extended version) | 94.4 | #9 | 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) |