AA-3D: Interactive 3D Experience Benchmark
AA-3D
AA-3D-Rangliste
Bereich | Modell | Elo | |
|---|---|---|---|
| 1 | 1-3 |  Claude Sonnet 5.5 (Max, Default Fallback) | 1194±3495%-Intervall 1160 bis 1228 |
| 2 | 1-3 | Claude Opus 5.5 (Max, Default Fallback) | 1190±3595%-Intervall 1155 bis 1225 |
| 3 | 1-4 |  GPT-6 Astra (Max) | 1165±3495%-Intervall 1131 bis 1199 |
| 4 | 4-6 | Claude Fable 5.1 (Max, Default Fallback) | 1131±3495%-Intervall 1098 bis 1165 |
| 5 | 4-6 | GPT-6 Sol (Max) | 1123±3495%-Intervall 1089 bis 1156 |
| 6 | 4-6 | GPT-6.1 Sol (Max) | 1114±3395%-Intervall 1081 bis 1148 |
| 7 | 7 |  Muse Spark 1.3 (Max) | 1035±3295%-Intervall 1003 bis 1067 |
| 8 | 8 | GPT-6 Luna (Max) | 1000AnkerAuf 1000 fixiert als Bezugspunkt für die Elo-Skala dieser Bestenliste. |
| 9 | 8-13 | Claude Haiku 5.5 (Max) | 994±3295%-Intervall 962 bis 1026 |
| 10 | 8-13 |  Grok 4.7 (Xhigh) | 994±3295%-Intervall 962 bis 1026 |
| 11 | 8-13 |  Kimi K3 (Max) | 982±3395%-Intervall 949 bis 1015 |
| 12 | 8-14 |  MiMo-V2.6-Pro | 977±3295%-Intervall 945 bis 1009 |
| 13 | 9-14 |  Gemini 3.8 Flash (High) | 965±3295%-Intervall 932 bis 997 |
| 14 | 12-15 |  Qwen3.8 Max (0902) | 948±3295%-Intervall 916 bis 981 |
| 15 | 14-15 |  GLM-5.3-Flash | 931±3295%-Intervall 899 bis 963 |
| 16 | 16-17 |  Mistral Large 4 Preview | 868±3295%-Intervall 835 bis 900 |
| 17 | 16-17 |  Step 5 Preview | 848±3395%-Intervall 814 bis 881 |
| 18 | 18 | GLM-5.3 (Max) | 782±3595%-Intervall 747 bis 816 |
| 19 | 19 |  MiniMax-M3 | 739±3495%-Intervall 704 bis 773 |
| 20 | 20-21 |  Inkling (Xhigh) | 579±4195%-Intervall 538 bis 620 |
| 21 | 20-21 |  Nemotron 3 Ultra 550B A55B (Reasoning) | 569±4595%-Intervall 524 bis 614 |
AA-3D Elo
AA-3D Elo
AA-3D: Elo vs. Kosten pro Aufgabe
Beispielaufgaben
Wähle einen Bereich, eine Aufgabe und ein Modell, um die Szene zu laden, die dieses Modell erstellt hat.
Aufgabe
Modell
Create an interactive adult man performing weightlifting exercises, with views that expose his muscles, skeleton and organs in anatomically accurate form.
21 von 21 Modellen haben eine Szene für diese Aufgabe in der Blindabstimmung.
Elo vs. Intelligence Index
AA-3D-Elo vs. Intelligence Index
Gewinnquote nach Bereich
Jede Zelle zeigt den Anteil der Stimmen, die ein Modell in einem Aufgabenbereich gewonnen hat, auf Basis der Stimmen hinter der aktuellen Rangliste. Die gegnerbereinigte Ansicht bildet den Mittelwert der Gewinnquoten des Modells gegen jeden Gegner im Bereich, sodass jeder Gegner gleich zählt.
AA-3D: Gewinnquote nach Bereich
Kosten
AA-3D: Kosten pro Aufgabe
Token-Nutzung
AA-3D: Ausgabe-Token pro Aufgabe
Turns pro Szene
AA-3D: Turns pro Szene
Werkzeugnutzung
AA-3D: Werkzeugaufrufe pro Szene
So funktioniert AA-3D
Jedes Modell arbeitet als Agent in einer Offline-Sandbox. Es erhält eine schriftliche Aufgabenbeschreibung für eine interaktive Szene und Abgabevorgaben, die für alle Aufgaben gelten. Das Modell schreibt Code, rendert die Szene, sieht sich die eigenen Screenshots an und überarbeitet sie. Zur Verfügung stehen Blender, Rapier-Physik, Python, ein Headless-Browser und eine Bibliothek mit CC0-Assets. Ein Lauf endet nach 3 Stunden oder 250 Turns. Das Ergebnis ist eine index.html-Seite, die die Szene im Browser ausführt.
Einblick in einen Agentenlauf
Diese Wiedergaben nutzen die Daten zum Start von AA-3D. Wähle eine Aufgabe, um den Lauf jedes Modells auf einer gemeinsamen Zeitachse abzuspielen. Fahre mit der Maus über eine Zeile, um jede Aktion zu sehen, und klicke darauf, um zu sehen, wohin die Zeit ging.
Sieh zu, wie eine Szene entsteht
Ergebnis vs. Veröffentlichungsdatum
AA-3D: Ergebnis vs. Veröffentlichungsdatum
Häufig gestellte Fragen
AA-3D v1.0 misst, wie gut KI-Modelle interaktive 3D-Erlebnisse erstellen. Jedes Modell erhält eine schriftliche Aufgabenbeschreibung und erstellt in einer Offline-Sandbox eine einzelne Webseite mit three.js. Der Benchmark umfasst 55 Aufgaben aus 7 Bereichen. AA-3D-Methodik lesen
Modelle werden mit einem Elo-Bewertungssystem gerankt, das aus Blindvergleichen unseres rekrutierten menschlichen Panels abgeleitet wird. Die Teilnehmenden interagieren mit zwei Szenen, die für dieselbe Aufgabe erstellt wurden, und wählen die bevorzugte, ohne zu wissen, welches Modell welche Szene erstellt hat. Höhere Elo-Werte bedeuten, dass ein Modell häufiger bevorzugt wird. AA-3D-Abstimmung ausprobieren
Evaluationen entdecken
Artificial Analysis Intelligence Index v4.3.2A composite benchmark aggregating ten challenging evaluations to provide a holistic measure of AI capabilities across mathematics, science, coding, and reasoning.
Artificial Analysis Openness IndexA composite measure providing an industry standard to communicate model openness for users and developers.
AA-Briefcase v1.1: Agentic Knowledge Work BenchmarkA private evaluation developed by Artificial Analysis for frontier agentic capability in long-horizon knowledge work, testing agents on realistic business workflows that require deliverables such as spreadsheets, presentations, and memos.
GDPval-AA v2.1 LeaderboardGDPval-AA v2.1 is Artificial Analysis' evaluation framework for OpenAI's GDPval dataset. It tests AI models on real-world tasks across 44 occupations and 9 major industries. Models are given shell access and web browsing capabilities in an agentic loop via Stirrup to solve tasks, with Elo ratings derived from blind pairwise comparisons.
AA-3D: Interactive 3D Experience BenchmarkA benchmark developed by Artificial Analysis for how well AI models build interactive 3D experiences. Models work as agents in an offline sandbox via Stirrup to turn a written brief into a three.js scene, with Elo ratings derived from blind pairwise votes.
APEX-Agents-AA Benchmark LeaderboardArtificial Analysis' implementation of the APEX-Agents benchmark, testing AI agents on long-horizon, cross-application tasks in professional-services environments with realistic application tooling.
AA-AnalystAgent Benchmark LeaderboardArtificial Analysis' data analysis benchmark, testing AI agents on their ability to work with spreadsheets and documents to answer quantitative questions a Business Analyst or Data Analyst would face day-to-day.
AutomationBench-AA: Agentic SaaS Workflow BenchmarkA benchmark measuring agentic task completion across simulated SaaS application environments, scoring the share of each task's objectives completed without guardrail violations.
Harvey LAB-AA v1.1 Benchmark LeaderboardArtificial Analysis' implementation of Harvey's Legal Agent Benchmark (LAB), testing AI agents on real-world legal work from Harvey's dataset of 120 private tasks spanning 24 legal practice areas. v1.1, built in collaboration with Harvey, checks every deliverable against the task's source documents for hallucinations, so work that passes every rubric criterion but contains a material hallucination receives no credit. A panel of three judges grades each deliverable criterion-by-criterion.
GDP.pdf Benchmark LeaderboardArtificial Analysis' implementation of Surge AI's GDP.pdf benchmark, testing whether language models can reason over long, real-world professional documents and satisfy detailed task-specific criteria.
EnterpriseOps-Gym-AA Benchmark LeaderboardArtificial Analysis' independent implementation of ServiceNow's EnterpriseOps-Gym, an agentic benchmark testing whether LLM agents can complete stateful, multi-step enterprise workflows across eight business domains via live tool use, graded on the final state of the underlying databases.
Terminal-Bench 4.0 Benchmark LeaderboardA harder 66-task benchmark of complex terminal work across software, machine learning, science, operations, security, hardware, and media, with recalibrated compute and time allowances and improved instructions, environments, and verifiers.
Terminal-Bench-Science 0.1 Benchmark LeaderboardA 70-task benchmark of research workflows authored and reviewed by domain experts across the life, physical, earth, mathematical, and engineering sciences, each completed in a terminal and checked by its own set of tests.
FrontierCode 1.1 Benchmark LeaderboardArtificial Analysis' implementation of Cognition's FrontierCode 1.1, a private agentic coding benchmark that tests models on completing coding tasks in a way that would actually get merged by maintainers of real-world repositories.
Artificial Analysis Long Context Reasoning Benchmark LeaderboardA challenging benchmark measuring language models' ability to extract, reason about, and synthesize information from long-form documents ranging from 10k to 100k tokens (measured using the cl100k_base tokenizer).
AA-Omniscience: Knowledge and Hallucination BenchmarkA benchmark measuring factual recall and hallucination across various economically relevant domains.
SciCode Benchmark LeaderboardA scientist-curated coding benchmark featuring 288 test set subproblems from 80 laboratory problems across 16 scientific disciplines.
Humanity's Last Exam Benchmark LeaderboardA frontier-level benchmark with 2,500 expert-vetted questions across mathematics, sciences, and humanities, designed to be the final closed-ended academic evaluation.
CritPt Benchmark LeaderboardA benchmark designed to test LLMs on research-level physics reasoning tasks, featuring 71 composite research challenges.
The most challenging 198 questions from GPQA, where PhD experts achieve 65% accuracy but skilled non-experts only reach 34% despite web access.
ITBench-AA Benchmark LeaderboardArtificial Analysis' implementation of IBM's ITBench benchmark, testing AI agents on Kubernetes incident root-cause analysis from offline incident snapshots. The agent inspects alerts, events, traces, and topology and identifies the contributing-factor entities (deployments, pods, namespaces, network policies, etc.) responsible for the failure.
MMMU-Pro Benchmark LeaderboardAn enhanced MMMU benchmark that eliminates shortcuts and guessing strategies to more rigorously test multimodal models across 30 academic disciplines.
IFBench Benchmark LeaderboardA benchmark evaluating precise instruction-following generalization on 58 diverse, verifiable out-of-domain constraints that test models' ability to follow specific output requirements.
Medical Long Context Reasoning (MLCR-AA)An open benchmark from Wisedocs measuring how well models reason over long, fragmented medical records, performing the multi-document synthesis claims professionals rely on when reviewing insurance and healthcare cases.
𝜏³-Banking Benchmark LeaderboardA fintech customer-support benchmark from the 𝜏-Knowledge framework that tests whether agents can navigate a large unstructured knowledge base and execute multi-step tool calls to resolve realistic banking workflows.
Terminal-Bench 2.1 Benchmark LeaderboardA verified refresh of Terminal-Bench 2.0 — 89 curated tasks across software engineering, system administration, data processing, model training, and security, with environment and instruction fixes so scores reflect agent capability rather than environment gaps.
Terminal-Bench Hard Benchmark LeaderboardAn agentic benchmark evaluating AI capabilities in terminal environments through software engineering, system administration, and data processing tasks.
𝜏²-Bench Telecom Benchmark LeaderboardA dual-control conversational AI benchmark simulating technical support scenarios where both agent and user must coordinate actions to resolve telecom service issues.
MMLU-Pro Benchmark LeaderboardAn enhanced version of MMLU with 12,000 graduate-level questions across 14 subject areas, featuring ten answer options and deeper reasoning requirements.
LiveCodeBench Benchmark LeaderboardA contamination-free coding benchmark that continuously harvests fresh competitive programming problems from LeetCode, AtCoder, and CodeForces, evaluating code generation, self-repair, and execution.
MATH-500 Benchmark LeaderboardA 500-problem subset from the MATH dataset, featuring competition-level mathematics across six domains including algebra, geometry, and number theory.
AIME 2025 Benchmark LeaderboardAll 30 problems from the 2025 American Invitational Mathematics Examination, testing olympiad-level mathematical reasoning with integer answers from 000-999.
Global-MMLU-Lite Benchmark LeaderboardA lightweight, multilingual version of MMLU, designed to evaluate knowledge and reasoning skills across a diverse range of languages and cultural contexts.