LLM-Capability-Matrix
Vergleiche Fähigkeiten und Parameter aktueller LLMs in einem interaktiven Vergleichsraster.
Loading…
Vergleiche Fähigkeiten und Parameter aktueller LLMs in einem interaktiven Vergleichsraster.
| Anbieter | Modell | Kontext | $/1M In ↑ | $/1M Out | Features | Tempo | Intelligenz | Ideal für |
|---|---|---|---|---|---|---|---|---|
| Anthropic | Claude Opus 4 claude-opus-4 | 200K | $15 | $75 | Vision-Support Function Calling JSON-Modus | mittel | State of the Art | SOTA reasoningExtended thinking+2 |
| Anthropic | Claude Sonnet 4 claude-sonnet-4 | 200K | $3 | $15 | Vision-Support Function Calling JSON-Modus | schnell | Sehr hoch | High-performanceBalanced tasks+2 |
| Anthropic | Claude Sonnet 3.7 claude-sonnet-3.7 | 200K | $3 | $15 | Vision-Support Function Calling JSON-Modus | mittel | Sehr hoch | Hybrid reasoningExtended thinking mode+1 |
| Anthropic | Claude Sonnet 3.5 claude-3.5-sonnet | 200K | $3 | $15 | Vision-Support Function Calling JSON-Modus | schnell | Sehr hoch | Balanced general-purposeCode generation+2 |
| Anthropic | Claude Haiku 3.5 claude-3.5-haiku | 200K | $0.8 | $4 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Fast chatLow-latency+2 |
| Anthropic | Claude Opus 3 claude-3-opus | 200K | $15 | $75 | Vision-Support Function Calling JSON-Modus | langsam | Sehr hoch | Complex reasoningResearch+2 |
| Anthropic | Claude Haiku 3 claude-3-haiku | 200K | $0.25 | $1.25 | Vision-Support Function Calling JSON-Modus | sehr schnell | Mittel | Ultra-low-costQuick tasks+2 |
| OpenAI | GPT-4.1 gpt-4.1 | 1.0M | $2 | $8 | Vision-Support Function Calling JSON-Modus | schnell | Sehr hoch | Flagship reasoningCoding+2 |
| OpenAI | GPT-4.1 mini gpt-4.1-mini | 1.0M | $0.4 | $1.6 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Balanced speed & qualityCost-effective+1 |
| OpenAI | GPT-4.1 nano gpt-4.1-nano | 1.0M | $0.1 | $0.4 | Function Calling JSON-Modus | ultraschnell | Mittel | Ultra-fastLow-cost+2 |
| OpenAI | GPT-4 Turbo 128k gpt-4-turbo | 128K | $10 | $30 | Vision-Support Function Calling JSON-Modus | mittel | Sehr hoch | High throughputComplex reasoning+1 |
| OpenAI | OpenAI o3 o3 | 200K | $10 | $40 | Vision-Support Function Calling JSON-Modus | mittel | State of the Art | Deep reasoningTool use+1 |
| OpenAI | OpenAI o4-mini o4-mini | 200K | $1.1 | $4.4 | Vision-Support Function Calling JSON-Modus | schnell | Hoch | Cost-efficient reasoningBalanced performance |
| OpenAI | GPT-4o gpt-4o | 128K | $5 | $20 | Vision-Support Function Calling JSON-Modus | schnell | Sehr hoch | Realtime multimodalVision+2 |
| OpenAI | GPT-4o mini gpt-4o-mini | 128K | $0.6 | $2.4 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Realtime multimodalCost-effective+1 |
| OpenAI | GPT-3.5 Turbo gpt-3.5-turbo-0125 | 16K | $0.5 | $1.5 | Function Calling JSON-Modus | schnell | Mittel | Low-cost chatSimple tasks+1 |
Gemini 2.5 Flash (Preview) gemini-2.5-flash | 1.0M | $0.15 | $0.6 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Hybrid reasoningLow-latency+2 | |
Gemini 2.5 Pro (Preview) gemini-2.5-pro | 1.0M | $1.25 | $10 | Vision-Support Function Calling JSON-Modus | mittel | Sehr hoch | Advanced codingReasoning+1 | |
Gemini 2.0 Flash gemini-2.0-flash | 1.0M | $0.1 | $0.4 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Balanced multimodalFast responses+1 | |
Gemini 2.0 Flash-Lite gemini-2.0-flash-lite | 1.0M | $0.075 | $0.3 | Vision-Support Function Calling JSON-Modus | ultraschnell | Mittel | Cost-optimizedHigh volume+1 | |
Gemini 1.5 Flash gemini-1.5-flash | 1.0M | $0.075 | $0.3 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Fastest GeminiMultimodal+2 | |
Gemini 1.5 Flash-8B gemini-1.5-flash-8b | 1.0M | $0.0375 | $0.15 | Vision-Support Function Calling JSON-Modus | ultraschnell | Mittel | Compact modelEdge deployment+1 | |
Gemini 1.5 Pro gemini-1.5-pro | 2.0M | $1.25 | $5 | Vision-Support Function Calling JSON-Modus | mittel | Sehr hoch | 2M token contextVideo analysis+2 | |
| xAI | Grok 3 grok-3 | 131K | $3 | $15 | Vision-Support Function Calling JSON-Modus | mittel | Sehr hoch | Strong reasoningX integration+1 |
| xAI | Grok 3 Fast grok-3-fast | 131K | $5 | $25 | Vision-Support Function Calling JSON-Modus | schnell | Sehr hoch | Low-latency GrokReal-time applications |
| xAI | Grok 3 Mini grok-3-mini | 131K | $0.3 | $0.5 | Vision-Support Function Calling JSON-Modus | schnell | Hoch | Smaller GrokCost-effective+1 |
| xAI | Grok 3 Mini Fast grok-3-mini-fast | 131K | $0.6 | $4 | Vision-Support Function Calling JSON-Modus | sehr schnell | Hoch | Low-latency miniFast responses+1 |
| xAI | Grok 2 1212 grok-2-1212 | 131K | $2 | $10 | Function Calling JSON-Modus | mittel | Hoch | Earlier GrokText-only tasks |
| xAI | Grok 2 Vision 1212 grok-2-vision-1212 | 8K | $2 | $10 | Vision-Support Function Calling JSON-Modus | mittel | Hoch | Vision-enabled Grok 2Image analysis |
| DeepSeek | DeepSeek Chat (V3) deepseek-chat | 66K | $0.27 | $1.1 | Function Calling JSON-Modus | schnell | Hoch | General chatJSON output+2 |
| DeepSeek | DeepSeek Reasoner (R1) deepseek-reasoner | 66K | $0.55 | $2.19 | Function Calling JSON-Modus | mittel | Sehr hoch | Chain-of-thought reasoningComplex problems+2 |
| Mistral | Mistral Large mistral-large | 33K | $8 | $24 | Function Calling JSON-Modus | mittel | Hoch | European languagesCode+2 |
| Mistral | Mixtral 8x7B mixtral-8x7b | 33K | $0.7 | $0.7 | JSON-Modus | schnell | Mittel-hoch | Cost effectiveOpen weights+2 |
| Cohere | Command R+ command-r-plus | 128K | $3 | $15 | Function Calling JSON-Modus | mittel | Hoch | RAGBusiness tasks+2 |
Kann Bilder und visuelle Inhalte verarbeiten
Kann Tools und API-Aufrufe ausführen
Garantiert strukturierte JSON-Ausgabe