The core LLMs and AI models powering the next generation of intelligent applications.
Perplexity AI is an answer engine that combines large language models with real-time web search to deliver accurate, cited responses.
Mistral AI builds high-performance open-weight language models that developers can deploy on their own infrastructure with full data control.
Cohere builds large language models and NLP APIs purpose-built for enterprise use cases with strong data privacy, security compliance and multilingual support.
Claude is Anthropic's flagship conversational AI assistant powered by the Claude 3 and Claude 3.5 model family, available via web, mobile, and API. Built with Constitutional AI principles, it emphasizes safety, honesty, and long-context comprehension across a wide range of professional and creative tasks.
GPT-4o is OpenAI's flagship omni model offering integrated text, image, and audio understanding in a single architecture. It is accessible via ChatGPT and the OpenAI API, serving both consumer and enterprise use cases at scale.
Llama 3 is Meta's open-weight large language model family released in 2024, available in 8B and 70B parameter sizes with instruction-tuned and base variants. It is freely downloadable for research and commercial use under Meta's custom community license.
DeepSeek V3 is a 671B Mixture-of-Experts open-weight foundational language model developed by DeepSeek AI, released in late 2024. It achieves top-tier benchmark performance across coding, reasoning, and general tasks while being freely available for download and self-deployment.
Gemini is Google's flagship family of multimodal large language models, available via gemini.google.com and the Gemini API, spanning Ultra, Pro, and Flash tiers. It supports text, image, audio, video, and code tasks with optional real-time Google Search integration.
Gemma 2 is Google DeepMind's family of open-weight language models ranging from 2B to 27B parameters, designed for research and production deployment with strong benchmark performance relative to model size. Released in 2024, it offers permissive licensing and broad hardware compatibility including consumer GPUs.
Qwen 2.5 is Alibaba Cloud's flagship large language model series offering strong multilingual reasoning, coding, and instruction-following capabilities across a range of parameter sizes. It is available via API at chat.qwenlm.ai and as open-weight models for self-hosted deployment.
Amazon Nova is a family of first-party foundational models available through Amazon Bedrock, spanning text, image, and video modalities across multiple capability tiers. Launched in late 2024, the Nova lineup ranges from lightweight edge-optimized models to premium multimodal reasoning models designed for enterprise-scale deployment.
Grok is xAI's flagship large language model, accessible via grok.x.ai and embedded within the X platform, featuring real-time web and X data access alongside multimodal capabilities. Built by Elon Musk's xAI, it competes directly with GPT-4 and Claude with an emphasis on wit, directness, and reduced content filtering.