Model Kataloğu
34 model · 7 provider · Fiyatlar 1M token başına USD
Anthropic's fastest model. Ideal for real-time interactions and summarization.
claude-3-haiku-20240307Anthropic's most powerful model. Best for highly complex tasks requiring deep expertise.
claude-3-opus-20240229Balanced intelligence and speed for enterprise workloads.
claude-3-sonnet-20240229Fast and affordable Claude with 200K context. Ideal for high-volume applications.
claude-3-5-haiku-20241022Anthropic's smartest model. Best for complex analysis, coding, and nuanced writing.
claude-3-5-sonnet-20241022Mistral model specialized for code generation and completion in 80+ languages.
codestral-latestClassic Cohere instruction-following model for text generation tasks.
commandOptimized for retrieval-augmented generation at lower cost.
command-rCohere's flagship model for RAG and enterprise search use cases.
command-r-plusChain-of-thought reasoning model. Competitive with o1 at a fraction of the price.
deepseek-reasonerDeepSeek R1 reasoning model distilled into Llama 70B, running on Groq LPU.
deepseek-r1-distill-llama-70bDeepSeek's best general-purpose model. Strong coding and reasoning at very low cost.
deepseek-chatFast, multimodal model with 1M context. Great price-performance ratio.
gemini-1.5-flashSmallest and fastest Gemini model. Ultra-low cost for high-volume use cases.
gemini-1.5-flash-8bMassive 2M token context window. Best for analyzing entire codebases or long documents.
gemini-1.5-proGoogle's next-gen multimodal model with 1M token context and native tool use.
gemini-2.0-flashGoogle's Gemma 2 9B model on Groq. Strong at reasoning and coding.
gemma2-9b-itFast and cost-effective. Great for simple tasks and chatbots.
gpt-3.5-turboThe original GPT-4. Highly capable for complex reasoning tasks.
gpt-4High-capability GPT-4 with 128K context and vision support.
gpt-4-turboOpenAI's flagship multimodal model. Fast and highly capable across text and vision tasks.
gpt-4oAffordable, intelligent, and fast. Best price-performance for most tasks.
gpt-4o-miniMeta Llama 3 70B on Groq. Strong performance for general tasks.
llama3-70b-8192Lightweight and fast Llama 3 model for quick responses.
llama3-8b-8192Meta Llama 3.1 70B on Groq LPU. Blazing fast inference with 131K context.
llama-3.1-70b-versatileExtremely fast and cheap. Best for real-time chat and simple completions.
llama-3.1-8b-instantMeta's latest Llama 3.3 70B model running on Groq's ultra-fast LPU inference.
llama-3.3-70b-versatileMistral's flagship large model. Strong at coding, reasoning, and multilingual tasks.
mistral-large-latestEfficient and affordable Mistral model for straightforward tasks.
mistral-small-latestSparse mixture-of-experts model. High capability with efficient inference.
mixtral-8x22b-instructPopular open-weight MoE model. Great balance of quality and speed.
mixtral-8x7b-instructOpenAI's reasoning model — thinks before responding. Excels at math, science, and complex logic.
o1Smaller, faster reasoning model. Best for STEM tasks at lower cost.
o1-miniLatest mini reasoning model. Excellent performance at a fraction of o1 cost.
o3-miniAuthorization: Bearer sk-... header POST /api/v1/chat/completions endpoint'ine istek gönder.