Modellvergleich: Analyse von Intelligenz, Leistung und Preis

Name: Intelligenz
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

Microevals Playground

Klicken Sie auf ein Modell, um detaillierte Metriken anzuzeigen. Weitere Informationen, auch zu unserer Methodik, finden Sie in unseren häufig gestellten Fragen.

Intelligenz

Updated

Claude Fable 5.1 (max with fallback) und

Claude Fable 5.1 (xhigh with fallback) sind die intelligentesten Modelle, gefolgt von

GPT-6 Astra (max) und

GPT-6 Astra (xhigh).

Ausgabegeschwindigkeit (Tokens/s)

Celeris-1 (1418 t/s) und

Mercury 2 (709 t/s) sind die schnellsten Modelle, gefolgt von

Gemini 2.5 Flash-Lite und

Gemini 3.5 Flash-Lite.

Latenz (Sekunden)

Gemini 2.5 Flash-Lite (Non-reasoning) (0.30s) und

Command A+ (0.38s) sind die Modelle mit der niedrigsten Latenz, gefolgt von

North Mini Code und

Gemini 2.5 Flash (Non-reasoning).

Preis ($ pro Mio. Tokens)

Devstral 2 ($0.00) und

North Mini Code ($0.00) sind die günstigsten Modelle, gefolgt von

Gemma 3 4B und

Gemma 3 27B.

Kontextfenster

Llama 4 Scout (10M) und

Grok 4.20 0309 (2M) sind die Modelle mit den größten Kontextfenstern, gefolgt von

Gemini 1.5 Pro (May) und

Grok 4.1 Fast.

Wichtigste Ergebnisse

Updated

Intelligenz

Artificial Analysis Intelligence Index · Higher is better

Geschwindigkeit

Output tokens per second · Higher is better

Kosten pro Aufgabe

Weighted average cost (USD) per Intelligence Index task · Lower is better

IntelligenzUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3 incorporates 10 evaluations: AA-Briefcase, GDPval-AA v2, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Artificial Analysis Intelligence Index v4.3 includes: AA-Briefcase, GDPval-AA v2, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Indicates whether the model weights are available. Models are labelled as 'Commercial Use Restricted' if commercial use is limited by conditions, and as 'Non-commercial' if the license prohibits commercial use.

FähigkeitsindizesUpdated

Misst die Leistung von Modellen bei bestimmten Fähigkeiten und in bestimmten Branchen

Artificial Analysis Finance & Accounting Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2, AA-Briefcase, Humanity's Last Exam, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Benchmarks

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

AA-Briefcase

Agentic knowledge work, (Elo-500)/2000

GDPval-AA v2

Agentic real-world work tasks, (Elo-500)/2000

AutomationBench-AAUpdated

Agentic SaaS workflows

Terminal-Bench 4.0New

Agentic coding & terminal use

SciCode

Coding

Humanity's Last Exam

Reasoning & knowledge

GDP.pdfNew

Professional document reasoning, All-pass

CritPt

Physics reasoning

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AA-LCR v1.1

Long context reasoning

Harvey LAB-AA

Legal agentic work, criterion pass rate

EnterpriseOps-Gym-AA

Agentic business operations

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

IFBench

Instruction following

𝜏³-Banking

Agentic tool use

APEX-Agents-AA

Long-horizon agentic tasks

ITBench-AA

Kubernetes incident root-cause analysis

MMMU-Pro

Visual reasoning

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

AA-Briefcase

AA-Briefcase Elo

AA-Briefcase is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better

AA-Briefcase Elo is a combined metric that aggregates analytical quality Elo, presentation Elo, and rubric pass rate, with rubric performance converted into Elo via synthetic head-to-head matches. Elo and 95% confidence interval bounds are clamped at 0.

AA-Omniscience

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct.

Openness Index

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open)

Vergleiche des Intelligence Index

Intelligence Index vs. Cost per Intelligence Index Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task

Most attractive quadrant

Pareto line

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

Tokenverbrauch

Output Tokens per Intelligence Index Task

Weighted average number of output tokens used to run one task in the Artificial Analysis Intelligence Index

The number of tokens required per Intelligence Index task. This is calculated by multiplying the output tokens per eval by the relative weights of each benchmark in the Intelligence Index, then dividing by task count (excluding repeats).

Kosten

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index

The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input, cache hit, cache write, reasoning, and answer token prices and the number of tokens used across evaluations (excluding repeats).

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens)

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Kontextfenster

Context Window

Context window: tokens limit · Higher is better

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Geschwindigkeit

Gemessen anhand der Ausgabegeschwindigkeit (Tokens pro Sekunde)

Output Speed

Output tokens per second · Higher is better

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Time per Intelligence Index Task

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better

The weighted average time (seconds) per Artificial Analysis Intelligence Index task. This is calculated by dividing output tokens per task by output speed, weighted by the relative weights of each benchmark in the Intelligence Index.

Latenz

Gemessen anhand der Zeit (in Sekunden) bis zum ersten Token

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time

Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.

Ende-zu-Ende-Antwortzeit

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

End-to-End Response Time

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better

Seconds to receive a 500 token response. Key components:

Input time: Time to receive the first response token
Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
Answer time: Time to generate 500 output tokens, based on output speed

Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).

Modellgröße (nur Modelle mit offenen Gewichten)

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

The number of parameters actually executed during each inference forward pass, expressed in billions. For Mixture of Experts (MoE) models, a routing mechanism selects a subset of experts per token, resulting in fewer active than total parameters. Dense models use all parameters, so active equals total.

Häufig gestellte Fragen

Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) führt derzeit den Artificial Analysis Intelligence Index mit einem Wert von 53 an. Insgesamt wurden 157 Modelle bewertet.

Die führenden KI-Modelle nach Intelligence Index sind: 1. Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) (53), 2. Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (53), 3. GPT-6 Astra (max) (53), 4. GPT-6 Astra (xhigh) (53) und 5. Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) (51).

Celeris-1 ist mit 1,418.0 Tokens pro Sekunde am schnellsten, gefolgt von Mercury 2 (708.9 Tokens/s) und Gemini 2.5 Flash-Lite (Reasoning) (416.3 Tokens/s).

Llama 3.1 Instruct 8B ist mit $0.02 pro 1 Mio. Tokens (Mischpreis) am günstigsten, gefolgt von Granite 4.2 3B ($0.02) und Nova Micro ($0.03).

Gemini 2.5 Flash-Lite (Non-reasoning) hat mit 0.30 s die kürzeste Zeit bis zum ersten Token, gefolgt von Command A+ (0.38 s) und North Mini Code (0.40 s).

GLM-5.3 (max) ist mit einem Intelligence-Index-Wert von 45 das am höchsten eingestufte Modell mit offenen Gewichten. Von insgesamt 157 bewerteten Modellen haben 78 Modelle offene Gewichte.

Die führenden KI-Modelle mit offenen Gewichten nach Intelligence Index sind: 1. GLM-5.3 (max) (45), 2. Kimi K3 (max) (44) und 3. GLM-5.3-Flash (42).

Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) führt unter 139 Reasoning-Modellen mit einem Intelligence-Index-Wert von 53. Reasoning-Modelle durchdenken komplexe Probleme ausführlich, bevor sie eine Antwort geben.

Modelle werden in mehreren Dimensionen verglichen, darunter Intelligenz (Qualität), Preis, Ausgabegeschwindigkeit (Tokens pro Sekunde), Latenz (Zeit bis zum ersten Token), Ende-zu-Ende-Antwortzeit und Größe des Kontextfensters. Die Leistungsmetriken werden mit standardisierten Prompts direkt über 651 Modelle gemessen.

Klicken Sie in den Diagrammen auf einen Modellnamen oder eine Zeile, um die zugehörige Seite mit detaillierten Metriken und direkten Vergleichen mit ähnlichen Modellen aufzurufen. Mit der Modellauswahl können Sie außerdem festlegen, welche Modelle in den einzelnen Diagrammen angezeigt werden. Bestenliste anzeigen

Modellvergleich: Analyse von Intelligenz, Leistung und Preis

IntelligenzUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index

Open Weights

FähigkeitsindizesUpdated

Intelligence Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

AA-Briefcase

AA-Briefcase Elo

AA-Briefcase Elo

AA-Omniscience

AA-Omniscience Index

AA-Omniscience Index

Openness Index

Artificial Analysis Openness Index: Score

Vergleiche des Intelligence Index

Intelligence Index vs. Cost per Intelligence Index Task

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

Tokenverbrauch

Output Tokens per Intelligence Index Task

Output Tokens per Intelligence Index Task

Kosten

Cost per Intelligence Index Task

Cost per Intelligence Index Task

Cost to Run Artificial Analysis Intelligence Index

Cost to Run Artificial Analysis Intelligence Index

Pricing: Cache Hit, Input, and Output

Cache Hit

Kontextfenster

Context Window

Context Window for RAG

Context Window

Geschwindigkeit

Output Speed

Output Speed

Model Performance Representation

Time per Intelligence Index Task

Time per Intelligence Index Task

Latenz

Latency: Time To First Answer Token

Time to First Answer Token

Ende-zu-Ende-Antwortzeit

End-to-End Response Time

End-to-End Response Time

Model Performance Representation

Modellgröße (nur Modelle mit offenen Gewichten)

Model Size: Total and Active Parameters

Total Parameters

Active Parameters at Inference Time

Häufig gestellte Fragen

Welches ist das intelligenteste KI-Modell?

Welches sind die besten KI-Modelle?

Welches ist das schnellste KI-Modell?

Welches ist das günstigste KI-Modell?

Welches KI-Modell hat die niedrigste Latenz?

Welches ist das beste KI-Modell mit offenen Gewichten?

Welches sind die besten KI-Modelle mit offenen Gewichten?

Welches ist das beste Reasoning-Modell?

Wie werden KI-Modelle auf Artificial Analysis verglichen?

Wie kann ich ein bestimmtes Modell mit anderen vergleichen?