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 3 · bench-align-v5
Gemini 4 Carbon is built for reasoning. It ranks in the top quarter for reasoning (top 2% of 216 scored models) and math (top 6% of 110 scored models), by BenchLM's published scores. It accepts text, images, PDFs, audio and video and produces text.
Gemini 4 Carbon by Google scores 88 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.
Every published price component, including cache reads and writes.
Lifecycle events the source associates with this model, plus what this profile does not claim.
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 | 75.02 s | 2026-11-21 | 2026-11-21 | |
| Throughput | 57 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 |
|---|---|---|---|---|---|
| Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index | 60.0 | #4 | Cross-benchmark intelligence index | Display-only external reference | Artificial Analysis |
| AA-Omniscience Index Artificial Analysis Omniscience Index | 53.5 | #3 | Knowledge questions | Broad factual knowledge | AA-Omniscience: Knowledge and Hallucination Benchmark |
| AA-Omniscience Accuracy Artificial Analysis Omniscience Accuracy | 58.3 | #11 | Knowledge questions | Broad knowledge | Artificial Analysis model benchmarks |
| AA-Omniscience Hallucination Rate Artificial Analysis Omniscience Hallucination Rate | 11.4 | #216 | Knowledge questions | Factuality | Artificial Analysis model benchmarks |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FrontierSWE v2 FrontierSWE v2 | 61.2 | #5 | 34 ultra-long-horizon engineering and research tasks | Ultra-long-horizon frontier software engineering | FrontierSWE v2 |
| Bug Hunt Bench Bug Hunt Bench | 57.3 | #3 | 105 planted bugs across two production TypeScript repositories | Blind production-repository bug finding and repair | Bug Hunt Bench method, caveats, and definitions |
| FrontierCode 2.0 FrontierCode 2.0 | 42.6 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3) FrontierMath v2 Tiers 1-3 | 90.526 | #7 | 285 private advanced mathematics problems | From olympiad-plus to early research mathematics | FrontierMath v2 benchmark hub |
| FrontierMath v2 (Tier 4) FrontierMath v2 Tier 4 | 95.122 | #5 | 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 | 97.50 | #5 | Semi-private ARC-AGI-1 evaluation set | Abstract visual reasoning | ARC Prize leaderboard |
| ARC-AGI-2 Abstraction and Reasoning Corpus for AGI v2 | 92.083333 | #5 | 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 | 63.855138 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AA Briefcase Artificial Analysis Briefcase | 1611 | #14 | Professional knowledge-work tasks | Professional work | Artificial Analysis Briefcase Benchmark Leaderboard |
| AA AutomationBench Artificial Analysis AutomationBench | 77.8 | #2 | Business-process automation tasks | Agentic automation | Artificial Analysis AutomationBench Benchmark Leaderboard |
| GDPval-AA GDPval-AA | 1733 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Chartography (no tools) Chartography without tools | 72.5 | #2 | 100 specialized chart types | Professional chart reasoning | Chartography |
| Design Arena Website Design Arena Website Elo | 1307 | #12 | Website generation comparisons | Design and website generation | OpenRouter Grok 4.3 benchmarks |
| Blueprint-Bench 2 Blueprint-Bench 2 | 58.6 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| SEC-Bench Pro SEC-Bench Pro | 88.2 | #2 | Security engineering tasks | Advanced cybersecurity | Introducing GPT-5.6 |
| Chess Puzzles Chess Puzzles (Epoch AI) | 78.0 | #2 | 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) | 73.0 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Gray Swan IPI (15 attempts) Gray Swan Indirect Prompt Injection, attack success at 15 attempts | 0.0 | #14 | 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 | 0.9653 ± 0.0121 | 2026-11-02 | |
| FrontierMath Tiers 1-3 (v2) Share of private research-level math problems solved, tiers 1-3. Unit: fraction (0–1); higher is better. | high | 0.9053 ± 0.0174 | 2026-11-02 | |
| FrontierMath Tier 4 (v2) Share of private research-level math problems solved, tier 4. Unit: fraction (0–1); higher is better. | high | 0.9512 ± 0.0341 | 2026-11-02 | |
| SimpleQA Verified Short factual questions answered correctly. Unit: fraction (0–1); higher is better. | high | 0.777 ± 0.0132 | 2026-11-02 |
Source: Epoch AI, CC BY 4.0 (licence), benchmark data read once a day.
| #7 |
| 132 private software-engineering tasks |
| Frontier coding-agent quality |
| FrontierCode 2.0 |
| PostTrainBench PostTrainBench v1.2 | 41.0 | #3 | Post-training four base models on six benchmarks | Autonomous ML engineering | PostTrainBench |
| #3 |
| 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 | 79.3 | #55 | Long-context reasoning tasks | Long-context reasoning | Artificial Analysis model benchmarks |
| MLCR-AA Medical Long Context Reasoning (MLCR-AA) | 19.4 | #29 | Long, fragmented medical-record reasoning | Long-context medical reasoning | Medical Long Context Reasoning (MLCR-AA) |
| GraphWalks BFS 256K–1M GraphWalks BFS 256K–1M | 96.2 | #1 | Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens. | Frontier agent evaluation | Gemini 4 Argon evaluation methodology |
| ARC-AGI-3 Provider Adapter ARC-AGI-3 Semi-Private: Provider Adapter harness | 96.85 | #5 | — | — | ARC Prize verified ARC-AGI-3 results |
| #15 |
| Agentic real-world work tasks |
| Professional agentic workflows |
| DeepSeek-V4 Technical Report |
| AA Terminal-Bench 4.0 Artificial Analysis Terminal-Bench v4.0 | 65.2 | #2 | Terminal-based agent tasks | Agentic software engineering | Artificial Analysis Terminal-Bench v4.0 Benchmark Leaderboard |
| GDP.pdf Artificial Analysis GDP.pdf | 28.4 | #9 | Professional document-production tasks | Professional knowledge work | GDP.pdf Benchmark Leaderboard |
| ExploitGym ExploitGym | 53.5 | #1 | 898 exploitation tasks | Advanced cybersecurity exploitation | ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? |
| AutomationBench AutomationBench | 53.9 | #5 | 600 public automation tasks | Long-horizon automation | Kimi K3: Open Frontier Intelligence |
| CWE-bench v1 CWE-bench v1 | 71.0 | #1 | 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 | 16 | #7 | 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 |
| AA-Briefcase (rubric) AA-Briefcase v1.1, rubric pass rate | 61.8 | #2 | Professional knowledge-work briefs | Professional knowledge work | Artificial Analysis AA-Briefcase |
| Terminal-Bench 4.1 Terminal-Bench 4.1 | 64.60 | #3 | 66 professional computer-work tasks | Frontier autonomous knowledge work | Terminal-Bench 4.1 |
| OSWorld 2.1 Offline OSWorld 2.1 offline subset (82 tasks) | 81.5 | #5 | 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 | 54.3 | #3 | 200 multimodal professional tasks across 10 industry verticals | Expert multimodal knowledge work | MKA Benchmark Leaderboard |
| #1 |
| Spatial reasoning from blueprints |
| Agentic spatial reasoning |
| Gemini 3.5 Flash launch screenshots |
| #6 |
| 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) | 77.5 | #7 | 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) | 87.6 | #3 | 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 |
| FACTS Multimodal FACTS Multimodal | 60.2 | #1 | About 1,500 image questions: 711 public and 811 held out | Image-grounded factual answering | BenchLM |
| DAYJOB DAYJOB | 19.7 | #3 | 130 long-horizon tasks: 80 finance and 50 healthcare | Long-horizon professional work | BenchLM |
| DAYJOB Healthcare DAYJOB Healthcare | 15.6 | #3 | 50 long-horizon healthcare tasks across five subfields | Long-horizon clinical and healthcare operations work | BenchLM |
| Landing Page Arena Landing Page Arena | 1463 | #9 | 12 client briefs for small-business landing pages, rated by human votes | Marketing page design and copy | BenchLM |
| Chart Arena Chart Arena | 1516 | #7 | 13 chart briefs with small datasets, rated by human votes | Data visualisation | BenchLM |
| Pelican Arena Pelican Arena | 1651 | #8 | 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 | 1574 | #6 | 20 famous landmarks built as three.js web pages, rated by human votes | 3D scenes in browser code | BenchLM |
| ASCII Arena ASCII Arena | 2020 | #1 | ASCII drawings of subjects from Google's Quick, Draw! dataset, rated by human votes | Text-character drawing | BenchLM |
| Game Arena Go Game Arena Go | 1830 | #2 | 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) | 99.3 | #1 | 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) |