Recommended Models
The following list contains the models we currently recommend for BYOK workspaces and the setup we use in our cloud (for non-BYOK workspaces). For the best user experience, we recommend to add at least the latest OpenAI GPT- and reasoning models, Claude models and Gemini models
This overview is only relevant for “Bring-your-own-key” (BYOK) customers of Langdock, who bring their own API Keys.
Minimum selection we recommend to allow users to use the strongest models:
| Model name | Hosting provider | Region availability | Image analysis | Supports Canvas | Supports Tools | Supports Reasoning | English text in app | German text in app | Important remarks | |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-4.1 | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | OpenAI’s flagship model. Good for most everyday tasks and for the data analyst (for CSV and Excel files). | OpenAI’s aktuelles Flagship-Modell. Gut für alltägliche Aufgaben und den Datenanalysten (für CSV und Excel Dateien). | ||
| GPT-4.1 nano | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smallest, fastest version of GPT-4.1. Ideal for instant answers, text completion, and tiny tasks. | Kleinste, schnellste Version von GPT-4.1. Ideal für sofortige Antworten, Textvervollständigung und kleine Aufgaben. | ||
| GPT-4.1 mini | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of GPT-4.1. Great for everyday tasks with significantly faster responses. | Kleinere, schnellere Version von GPT-4.1. Geeignet für alltägliche Aufgaben mit deutlich schnelleren Antworten. | ||
| o3 | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Reasoning model from OpenAI. Takes time to think deeply for superior coding, math, science, and visual analysis capabilities. | OpenAI’s Flagship Reasoning-Modell. Denkt mehr nach für überlegene Fähigkeiten in Coding, Mathe, Wissenschaft und visueller Analyse. | Please select “Responses API” to display reasoning in model response. | |
| o4 mini | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Small reasoning model optimized for speed. Exceptional performance in math, coding, and visual analysis tasks. | Neueste Version der kleinen, schnellen Reasoning-Modelle von OpenAI. Geeignet für komplexe Aufgaben. | Please select “Responses API” to display reasoning in model response. | |
| Claude Sonnet 4 | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Anthropic’s flagship model for complex tasks. Excels at coding, authentic human-sounding text, and following detailed instructions. | Anthropic’s Flagship-Modell für komplexe Aufgaben. Hervorragend in Coding, authentischen Texten und beim befolgen detaillierten Anweisungen. | ||
| Claude Sonnet 4 Reasoning | Google Vertex | EU/US | ✅ | ✅ | ✅ | ✅ | Extended thinking version of Claude Sonnet 4. Ideal for complex problems requiring step-by-step analysis. | Reasoning-Version von Claude Sonnet 4. Ideal für komplexe Probleme mit Chain-of-Thought-Analyse. | ||
| Gemini 2.5 Pro | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Google’s flagship model for complex tasks. Uses enhanced reasoning and excels at math, coding, and analyzing large datasets. | Google’s Flagship-Modell für komplexe Aufgaben. Nutzt verbessertes Reasoning, hervorragend in Mathe, Coding und großen Datensätzen. | ||
| Gemini 2.5 Flash | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of Gemini 2.5 Pro. Great balance of speed and intelligence for high-volume everyday tasks. | Kleinere, schnellere Version von Gemini 2.5 Pro. Gute Balance aus Geschwindigkeit und Intelligenz für alltägliche High-Volume-Aufgaben. |
Additional models that are supported:
| Model name | Hosting provider | Region availability | Important remarks | Image analysis | Supports Canvas | Supports Tools | Supports Reasoning | English text in app | German text in app |
|---|---|---|---|---|---|---|---|---|---|
| GPT-4o | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Versatile, intelligent model for diverse tasks. Excels at reasoning, coding, and understanding complex content. | Vielseitiges, intelligentes Modell für vielfältige Aufgaben. Exzellent im Argumentieren, Coden und Verstehen komplexer Inhalte. | |
| GPT-4o mini | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of GPT-4o. Great balance of intelligence and speed for smaller tasks. | Kleinere, schnellere Version von GPT-4o. Gute Balance aus Intelligenz und Geschwindigkeit für kleinere Aufgaben. | |
| GPT-4.5 | Microsoft Azure | Global/US | ✅ | ❌ | ✅ | ❌ | This model will be deprecated on the 11th of July. We recommend switching to GPT-4.1 | Dieses Modell wird am 11. Juli eingestellt. Wir empfehlen, zu GPT-4.1 zu wechseln. | |
| o1 | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Advanced reasoning model that thinks step-by-step. Excels at complex math, science, and coding problems requiring deep analysis. | Logik-Modell, das Schritt für Schritt vorgeht. Geeignet für komplexe Mathematik-, Wissenschafts- und Programmieraufgaben. | |
| o3 Mini | Microsoft Azure | EU/US | To use the standard mode, enter “o3-mini” as the model id in Langdock. | ✅ | ❌ | ✅ | ✅ | Smaller, faster version of o3. Enhanced reasoning with significantly faster responses. | Kleinere, schnellere Version von o3. Verbessertes Reasoning mit deutlich schnelleren Antworten. |
| o3 Mini high | Microsoft Azure | EU/US | To use the high reasoning mode, enter “o3-mini-high” as the model id in Langdock. | ✅ | ❌ | ✅ | ✅ | Higher-intelligence version of o3 mini. Enhanced reasoning power for complex math, science, and coding problems. | Version mit höherer Denkkapazität von o3 mini. Verstärktes Reasoning für komplexe Probleme. |
| Claude Sonnet 3.7 | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Intelligent model with strong coding and analysis abilities. Excels at complex problem-solving and detailed tasks. | Intelligentes Modell mit starken Coding- und Analyse-Fähigkeiten. Hervorragend in komplexer Problemlösung und detaillierten Aufgaben. | |
| Claude Sonnet 3.7 Reasoning | Google Vertex | EU/US | ✅ | ✅ | ✅ | ✅ | Extended thinking version of Claude Sonnet 3.7. Deep analytical thinking for complex math, coding, and logic problems. | Erweiterte Thinking-Version von Claude Sonnet 3.7. Tiefes analytisches Denken für komplexe Mathe-, Coding- und Logik-Probleme. | |
| DeepSeek r1 | Microsoft Azure | US | ❌ | ❌ | ❌ | ✅ | Open-source reasoning model with strong analytical capabilities. Excels at math, coding, and step-by-step problem solving. | Open-Source Reasoning-Modell mit starken analytischen Fähigkeiten. Hervorragend in Mathe, Coding und schrittweiser Problemlösung. | |
| Llama 3.3 70B | Microsoft Azure | EU/US | ❌ | ❌ | ✅ | ❌ | Flagship open-source model by Meta. Good at text tasks and basic problem-solving. Reliable option for everyday activities. | Open-Source-Modell von Meta. Gut in Text-Aufgaben und grundlegender Problemlösung. Zuverlässig für alltägliche Aktivitäten. | |
| Mistral Large 2411 | Microsoft Azure | EU/US | ❌ | ❌ | ✅ | ❌ | Advanced model with strong coding abilities. Excels at complex problem-solving, math, and multilingual tasks. | Fortgeschrittenes Modell mit starken Coding-Fähigkeiten. Hervorragend in komplexer Problemlösung, Mathe und mehrsprachigen Aufgaben. | |
| Sonar, Sonar Pro, Sonar Reasoning | Perplexity | US | Not offered in our Cloud at the moment | ❌ | ❌ | ❌ | ❌ | n.a. | n.a. |
| Amazon Nova Lite | AWS | US | ❌ | ❌ | ❌ | ❌ | Smaller version of Amazon’s current flagship model. | Kleinere Version von Amazons aktuellem Flaggschiff-Modell. | |
| Amazon Nova Pro | AWS | US | ❌ | ❌ | ❌ | ❌ | Amazon’s flagship multimodal model. Strong at reasoning, coding, and analyzing diverse content. | Amazons multimodales Flaggschiff-Modell. Stark in logischem Denken, Programmierung und der Analyse vielfältiger Inhalte. | |
| Flux Black Forest Labs | Black Forest Labs / Replicate | US | Not offered in our Cloud at the moment | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. |
Recommended Models
The following list contains the models we currently recommend for BYOK workspaces and the setup we use in our cloud (for non-BYOK workspaces). For the best user experience, we recommend to add at least the latest OpenAI GPT- and reasoning models, Claude models and Gemini models
This overview is only relevant for “Bring-your-own-key” (BYOK) customers of Langdock, who bring their own API Keys.
Minimum selection we recommend to allow users to use the strongest models:
| Model name | Hosting provider | Region availability | Image analysis | Supports Canvas | Supports Tools | Supports Reasoning | English text in app | German text in app | Important remarks | |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-4.1 | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | OpenAI’s flagship model. Good for most everyday tasks and for the data analyst (for CSV and Excel files). | OpenAI’s aktuelles Flagship-Modell. Gut für alltägliche Aufgaben und den Datenanalysten (für CSV und Excel Dateien). | ||
| GPT-4.1 nano | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smallest, fastest version of GPT-4.1. Ideal for instant answers, text completion, and tiny tasks. | Kleinste, schnellste Version von GPT-4.1. Ideal für sofortige Antworten, Textvervollständigung und kleine Aufgaben. | ||
| GPT-4.1 mini | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of GPT-4.1. Great for everyday tasks with significantly faster responses. | Kleinere, schnellere Version von GPT-4.1. Geeignet für alltägliche Aufgaben mit deutlich schnelleren Antworten. | ||
| o3 | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Reasoning model from OpenAI. Takes time to think deeply for superior coding, math, science, and visual analysis capabilities. | OpenAI’s Flagship Reasoning-Modell. Denkt mehr nach für überlegene Fähigkeiten in Coding, Mathe, Wissenschaft und visueller Analyse. | Please select “Responses API” to display reasoning in model response. | |
| o4 mini | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Small reasoning model optimized for speed. Exceptional performance in math, coding, and visual analysis tasks. | Neueste Version der kleinen, schnellen Reasoning-Modelle von OpenAI. Geeignet für komplexe Aufgaben. | Please select “Responses API” to display reasoning in model response. | |
| Claude Sonnet 4 | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Anthropic’s flagship model for complex tasks. Excels at coding, authentic human-sounding text, and following detailed instructions. | Anthropic’s Flagship-Modell für komplexe Aufgaben. Hervorragend in Coding, authentischen Texten und beim befolgen detaillierten Anweisungen. | ||
| Claude Sonnet 4 Reasoning | Google Vertex | EU/US | ✅ | ✅ | ✅ | ✅ | Extended thinking version of Claude Sonnet 4. Ideal for complex problems requiring step-by-step analysis. | Reasoning-Version von Claude Sonnet 4. Ideal für komplexe Probleme mit Chain-of-Thought-Analyse. | ||
| Gemini 2.5 Pro | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Google’s flagship model for complex tasks. Uses enhanced reasoning and excels at math, coding, and analyzing large datasets. | Google’s Flagship-Modell für komplexe Aufgaben. Nutzt verbessertes Reasoning, hervorragend in Mathe, Coding und großen Datensätzen. | ||
| Gemini 2.5 Flash | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of Gemini 2.5 Pro. Great balance of speed and intelligence for high-volume everyday tasks. | Kleinere, schnellere Version von Gemini 2.5 Pro. Gute Balance aus Geschwindigkeit und Intelligenz für alltägliche High-Volume-Aufgaben. |
Additional models that are supported:
| Model name | Hosting provider | Region availability | Important remarks | Image analysis | Supports Canvas | Supports Tools | Supports Reasoning | English text in app | German text in app |
|---|---|---|---|---|---|---|---|---|---|
| GPT-4o | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Versatile, intelligent model for diverse tasks. Excels at reasoning, coding, and understanding complex content. | Vielseitiges, intelligentes Modell für vielfältige Aufgaben. Exzellent im Argumentieren, Coden und Verstehen komplexer Inhalte. | |
| GPT-4o mini | Microsoft Azure | EU/US | ✅ | ✅ | ✅ | ❌ | Smaller, faster version of GPT-4o. Great balance of intelligence and speed for smaller tasks. | Kleinere, schnellere Version von GPT-4o. Gute Balance aus Intelligenz und Geschwindigkeit für kleinere Aufgaben. | |
| GPT-4.5 | Microsoft Azure | Global/US | ✅ | ❌ | ✅ | ❌ | This model will be deprecated on the 11th of July. We recommend switching to GPT-4.1 | Dieses Modell wird am 11. Juli eingestellt. Wir empfehlen, zu GPT-4.1 zu wechseln. | |
| o1 | Microsoft Azure | EU/US | ✅ | ❌ | ✅ | ✅ | Advanced reasoning model that thinks step-by-step. Excels at complex math, science, and coding problems requiring deep analysis. | Logik-Modell, das Schritt für Schritt vorgeht. Geeignet für komplexe Mathematik-, Wissenschafts- und Programmieraufgaben. | |
| o3 Mini | Microsoft Azure | EU/US | To use the standard mode, enter “o3-mini” as the model id in Langdock. | ✅ | ❌ | ✅ | ✅ | Smaller, faster version of o3. Enhanced reasoning with significantly faster responses. | Kleinere, schnellere Version von o3. Verbessertes Reasoning mit deutlich schnelleren Antworten. |
| o3 Mini high | Microsoft Azure | EU/US | To use the high reasoning mode, enter “o3-mini-high” as the model id in Langdock. | ✅ | ❌ | ✅ | ✅ | Higher-intelligence version of o3 mini. Enhanced reasoning power for complex math, science, and coding problems. | Version mit höherer Denkkapazität von o3 mini. Verstärktes Reasoning für komplexe Probleme. |
| Claude Sonnet 3.7 | Google Vertex | EU/US | ✅ | ✅ | ✅ | ❌ | Intelligent model with strong coding and analysis abilities. Excels at complex problem-solving and detailed tasks. | Intelligentes Modell mit starken Coding- und Analyse-Fähigkeiten. Hervorragend in komplexer Problemlösung und detaillierten Aufgaben. | |
| Claude Sonnet 3.7 Reasoning | Google Vertex | EU/US | ✅ | ✅ | ✅ | ✅ | Extended thinking version of Claude Sonnet 3.7. Deep analytical thinking for complex math, coding, and logic problems. | Erweiterte Thinking-Version von Claude Sonnet 3.7. Tiefes analytisches Denken für komplexe Mathe-, Coding- und Logik-Probleme. | |
| DeepSeek r1 | Microsoft Azure | US | ❌ | ❌ | ❌ | ✅ | Open-source reasoning model with strong analytical capabilities. Excels at math, coding, and step-by-step problem solving. | Open-Source Reasoning-Modell mit starken analytischen Fähigkeiten. Hervorragend in Mathe, Coding und schrittweiser Problemlösung. | |
| Llama 3.3 70B | Microsoft Azure | EU/US | ❌ | ❌ | ✅ | ❌ | Flagship open-source model by Meta. Good at text tasks and basic problem-solving. Reliable option for everyday activities. | Open-Source-Modell von Meta. Gut in Text-Aufgaben und grundlegender Problemlösung. Zuverlässig für alltägliche Aktivitäten. | |
| Mistral Large 2411 | Microsoft Azure | EU/US | ❌ | ❌ | ✅ | ❌ | Advanced model with strong coding abilities. Excels at complex problem-solving, math, and multilingual tasks. | Fortgeschrittenes Modell mit starken Coding-Fähigkeiten. Hervorragend in komplexer Problemlösung, Mathe und mehrsprachigen Aufgaben. | |
| Sonar, Sonar Pro, Sonar Reasoning | Perplexity | US | Not offered in our Cloud at the moment | ❌ | ❌ | ❌ | ❌ | n.a. | n.a. |
| Amazon Nova Lite | AWS | US | ❌ | ❌ | ❌ | ❌ | Smaller version of Amazon’s current flagship model. | Kleinere Version von Amazons aktuellem Flaggschiff-Modell. | |
| Amazon Nova Pro | AWS | US | ❌ | ❌ | ❌ | ❌ | Amazon’s flagship multimodal model. Strong at reasoning, coding, and analyzing diverse content. | Amazons multimodales Flaggschiff-Modell. Stark in logischem Denken, Programmierung und der Analyse vielfältiger Inhalte. | |
| Flux Black Forest Labs | Black Forest Labs / Replicate | US | Not offered in our Cloud at the moment | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. |