Name
Summary
Quality
Speed
Cost
Hosting
Anthropic
Anthropic
/
Claude 4 Sonnet
Claude 4 Sonnet
Versteht Nuancen, analysiert Daten, schreibt Inhalte.
Quality
80%
Speed
24%
Cost
100%
Hosting Location
Dieses Modell versteht subtile Anweisungen und Kontext. Es analysiert komplexe Informationen und erstellt nuancierten Text. Es bietet ein großes Kontextfenster und geringe Raten fehlerhafter Ausgaben.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Anthropic
Version
-
Languages
0
Type
Chat
License
Proprietary
Hoster
AWS Bedrock
Hoster Company Location
USA
Hosting Location
Frankfurt, Deutschland
Knowledge Cutoff
March 2025
Context-Window
200000
Token Input
-
Token Output
-
Purpose:
Hosted by AWS
8 Sprachen
Multilingual
Anthropic
/
Claude 3.7 Sonnet
Claude 3.7 Sonnet
Bietet schnelle Antworten oder sichtbares Schritt-für-Schritt-Denken.
Quality
73%
Speed
15%
Cost
0%
Hosting Location
Bietet schnelle Antworten für schnelle Anforderungen. Für komplexere Aufgaben kann es einen erweiterten, sichtbaren Schritt-für-Schritt-Denkprozess nutzen. Dies hilft beim Verständnis und Befolgen von Anweisungen. Das Modell ist für praktische Anwendungen trainiert.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Anthropic
Version
-
Languages
0
Type
Chat
License
Proprietary
Hoster
AWS Bedrock
Hoster Company Location
USA
Hosting Location
Frankfurt, Deutschland
Knowledge Cutoff
October 2024
Context-Window
200000
Token Input
-
Token Output
-
Purpose:
Hosted by AWS
8 Sprachen
Multilingual
Anthropic
/
Claude 3.5 Sonnet
Claude 3.5 Sonnet
Versteht Nuancen, folgt komplexen Textanweisungen gut.
Quality
83%
Speed
27%
Cost
0%
Hosting Location
Dieses Modell erfasst Gesprächsverläufe und detaillierte Textanweisungen präzise. Es liefert zusammenhängende, nützliche Textantworten und behält Kontext in längeren Unterhaltungen bei.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Anthropic
Version
-
Languages
18
Type
Chat
License
Proprietary
Hoster
AWS Bedrock
Hoster Company Location
USA
Hosting Location
Frankfurt, Deutschland
Knowledge Cutoff
April 2024
Context-Window
200000
Token Input
3.17€/1M Tokens
Token Output
15.84€/1M Tokens
Purpose:
Hosted by AWS
Logisches Denken
Multilingual
Cohere
Cohere
/
Command R+
Command R+
Entwickelt für die Bedürfnisse von Unternehmen.
Quality
32%
Speed
18%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Cohere
Version
08-2024
Languages
0
Type
Chat
License
Proprietary
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
February 2023
Context-Window
128000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
Cohere
/
Command R
Command R
Ein einfaches Modell für die Dokumentensuche und Verarbeitung.
Quality
22%
Speed
41%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Cohere
Version
08-2024
Languages
0
Type
Chat
License
Proprietary
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
February 2023
Context-Window
128000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
Cohere
/
Command A
Command A
Schnelle Textantworten, verarbeitet lange Eingaben, viele Sprachen.
Quality
60%
Speed
60%
Cost
0%
Hosting Location
Dieses Modell generiert Textantworten schnell und effizient. Es verarbeitet sehr lange Texteingaben und gibt präzise Antworten, auch basierend auf externen Informationen. Es funktioniert gut im Chat und unterstützt dabei viele verschiedene Sprachen zuverlässig.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Cohere
Version
-
Languages
0
Type
Instruct
License
Proprietary
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
June 2024
Context-Window
256000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
Deepseek
Deepseek
/
DeepSeek-V3
DeepSeek-V3
Hervorragend in der Erstellung kohärenter, inhaltsstarker Texte.
Quality
68%
Speed
27%
Cost
34%
Hosting Location
-
Model Origin
China
Hoster Company Location
USA
Hosting Location
USA
Creator
Deepseek
Version
-
Languages
0
Type
Chat
License
OpenSource
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
June 2024
Context-Window
128000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Open source
Multilingual
Deepseek
/
DeepSeek-R1
DeepSeek-R1
Das erste starke Open Source Reasoning Modell.
Quality
100%
Speed
14%
Cost
44%
Hosting Location
-
Model Origin
China
Hoster Company Location
USA
Hosting Location
USA
Creator
Deepseek
Version
-
Languages
0
Type
Instruct
License
OpenSource
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
June 2024
Context-Window
128000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Open source
Multilingual
Google
Google
/
Gemini 2.5 Flash
Gemini 2.5 Flash
Das Flaggschiff-Modell von Google - intelligent und schnell.
Quality
80%
Speed
84%
Cost
4%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Google
Version
preview-05-20
Languages
0
Type
Chat
License
Proprietary
Hoster
Google Vertex AI
Hoster Company Location
USA
Hosting Location
Eemshaven, Niederlande
Knowledge Cutoff
January 2025
Context-Window
1048576
Token Input
-
Token Output
-
Purpose:
Logisches Denken
Hosted by Google
Multilingual
Google
/
Gemini 2.0 Flash-Lite
Gemini 2.0 Flash-Lite
Ein günstiges und schnelles Modell von Google.
Quality
63%
Speed
84%
Cost
2%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Google
Version
001
Languages
0
Type
Chat
License
Proprietary
Hoster
Google Vertex AI
Hoster Company Location
USA
Hosting Location
Eemshaven, Niederlande
Knowledge Cutoff
June 2024
Context-Window
1048576
Token Input
-
Token Output
-
Purpose:
Logisches Denken
Hosted by Google
Multilingual
Google
/
Gemini 2.0 Flash
Gemini 2.0 Flash
Bietet ultraschnelle Antworten und hohe Qualität mit großem Kontext.
Quality
100%
Speed
92%
Cost
3%
Hosting Location
Gemini 2.0 Flash bietet Funktionen der nächsten Generation und verbesserte Funktionen, darunter eine höhere Geschwindigkeit, die Nutzung nativer Tools, die multimodale Generierung und ein Kontextfenster von 1 Million Tokens.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Google
Version
001
Languages
46
Type
Chat
License
Proprietary
Hoster
Google Vertex AI
Hoster Company Location
USA
Hosting Location
Eemshaven, Niederlande
Knowledge Cutoff
August 2024
Context-Window
1000000
Token Input
-
Token Output
-
Purpose:
Hosted by Google
Logisches Denken
Multilingual
Google
/
Gemini 1.5 Pro
Gemini 1.5 Pro
Für kostengünstige Anwendungen mit hohem Volumen entwickelt.
Quality
67%
Speed
35%
Cost
37%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Google
Version
002
Languages
0
Type
Chat
License
Proprietary
Hoster
Google Vertex AI
Hoster Company Location
USA
Hosting Location
Eemshaven, Niederlande
Knowledge Cutoff
May 2024
Context-Window
1000000
Token Input
-
Token Output
-
Purpose:
Logisches Denken
Hosted by Google
Multilingual
Google
/
Gemini 1.5 Flash
Gemini 1.5 Flash
Ein leichtgewichtiges Modell für Geschwindigkeit und Effizienz.
Quality
59%
Speed
71%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
Google
Version
002
Languages
0
Type
Chat
License
Proprietary
Hoster
Google Vertex AI
Hoster Company Location
USA
Hosting Location
Eemshaven, Niederlande
Knowledge Cutoff
May 2024
Context-Window
1048576
Token Input
-
Token Output
-
Purpose:
Logisches Denken
Hosted by Google
Multilingual
Meta
Meta
/
Llama 4 Scout 17B
Llama 4 Scout 17B
Versteht lange Texte in diversen Sprachen
Quality
65%
Speed
46%
Cost
0%
Hosting Location
Verarbeitet sehr lange Textgespräche bis zu 10 Millionen Tokens und fasst Informationen zusammen. Das Modell antwortet in 12 Sprachen.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Meta
Version
17B-16E-Instruct
Languages
12
Type
Instruct
License
OpenSource
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
August 2024
Context-Window
10000000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Open source
Multilingual
Meta
/
Llama 4 Maverick 17B
Llama 4 Maverick 17B
Verarbeitet lange Texte, liefert hochwertige Antworten, unterstützt viele Sprachen.
Quality
76%
Speed
60%
Cost
0%
Hosting Location
Dieses Modell kann lange Texteingaben gut verarbeiten und behält dabei den Kontext. Es liefert qualitativ gute und ausgeglichene Antworten für allgemeine Unterhaltungen und logische Aufgaben. Das Modell unterstützt mehrere Sprachen.
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Meta
Version
17B-128E-Instruct-FP8
Languages
12
Type
Instruct
License
Proprietary
Hoster
Azure AI Foundry
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
August 2024
Context-Window
1000000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Open source
Multilingual
Meta
/
Llama 3.3 70B Instruct
Llama 3.3 70B Instruct
Optimiert für Chat und flexibel für verschiedene Sprachaufgaben.
Quality
85%
Speed
100%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
USA
Creator
Meta
Version
-
Languages
8
Type
Instruct
License
OpenSource
Hoster
AWS Bedrock
Hoster Company Location
USA
Hosting Location
Global
Knowledge Cutoff
December 2023
Context-Window
128000
Token Input
0.76€/1M Tokens
Token Output
0.76€/1M Tokens
Purpose:
Hosted by AWS
Open source
8 Sprachen
Mistral AI
Mistral AI
/
Mistral Large
Mistral Large
Präzises Modell für mehrsprachiges Reasoning und Code.
Quality
55%
Speed
16%
Cost
0%
Hosting Location
-
Model Origin
European Union
Hoster Company Location
USA
Hosting Location
European Union
Creator
Mistral AI
Version
-
Languages
5
Type
Chat
License
OpenSource
Hoster
AWS Bedrock
Hoster Company Location
USA
Hosting Location
Dublin, Irland
Knowledge Cutoff
October 2023
Context-Window
32000
Token Input
4.54€/1M Tokens
Token Output
13.73€/1M Tokens
Purpose:
Hosted by AWS
Open source
5 Sprachen
OpenAI
OpenAI
/
o3-mini
o3-mini
Ein leichtgewichtiges Reasoning Modell für komplexe Fragestellungen.
Quality
95%
Speed
77%
Cost
33%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
OpenAI
Version
-
Languages
0
Type
Chat
License
Proprietary
Hoster
Azure OpenAI
Hoster Company Location
USA
Hosting Location
Gävle, Schweden
Knowledge Cutoff
October 2023
Context-Window
4000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
OpenAI
/
gpt-4o-mini
gpt-4o-mini
Schnelles, günstiges Modell für fokussierte Aufgaben
Quality
74%
Speed
67%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
OpenAI
Version
2024-07-18
Languages
74
Type
Chat
License
Proprietary
Hoster
Azure OpenAI
Hoster Company Location
USA
Hosting Location
Gävle, Schweden
Knowledge Cutoff
October 2023
Context-Window
128000
Token Input
0.16€/1M Tokens
Token Output
0.633€/1M Tokens
Purpose:
Hosted by Azure
Multilingual
Logisches Denken
OpenAI
/
GPT-4o
GPT-4o
Ein besonders starkes Modell für komplexe Fragestellungen
Quality
86%
Speed
27%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
OpenAI
Version
2024-11-20
Languages
74
Type
Chat
License
Proprietary
Hoster
Azure OpenAI
Hoster Company Location
USA
Hosting Location
Gävle, Schweden
Knowledge Cutoff
October 2023
Context-Window
128000
Token Input
2.64€/1M Tokens
Token Output
10.55€/1M Tokens
Purpose:
Hosted by Azure
Multilingual
Logisches Denken
OpenAI
/
GPT 4.1 nano
GPT 4.1 nano
Ein kleines, auf Schnelligkeit optimiertes Modell.
Quality
62%
Speed
93%
Cost
0%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
OpenAI
Version
-
Languages
0
Type
Chat
License
Proprietary
Hoster
Azure OpenAI
Hoster Company Location
USA
Hosting Location
Gävle, Schweden
Knowledge Cutoff
June 2024
Context-Window
1000000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
OpenAI
/
GPT 4.1
GPT 4.1
Besonders stark bei langen Kontexten und im befolgen von Anweisungen.
Quality
79%
Speed
78%
Cost
51%
Hosting Location
-
Model Origin
USA
Hoster Company Location
USA
Hosting Location
European Union
Creator
OpenAI
Version
-
Languages
0
Type
Chat
License
Proprietary
Hoster
Azure OpenAI
Hoster Company Location
USA
Hosting Location
Gävle, Schweden
Knowledge Cutoff
June 2024
Context-Window
1000000
Token Input
-
Token Output
-
Purpose:
Hosted by Azure
Multilingual
Enterprise-grade security
ISO 27001 (coming 2025)
KI Made in Germany
Highest standards. Clear values.
GDPR and AI Act compliant
Hosted in Europe