Skip to content
  • There are no suggestions because the search field is empty.

Overview

Compass is a comprehensive generative AI enablement platform designed to empower companies to harness AI's transformative power. Powered by Core42, Compass offers a robust suite of APIs using leading AI models like GPT-5, GPT-5 mini, GPT-4.1, GPT-4o, GPT-4o mini, GPT-4o Audio, GPT Image 1, o3, o3-mini, o4-mini, gpt-oss-120b, gpt-oss-20b, GPT-4o mini TTS, GPT-4o Transcribe, GPT-4o mini Transcribe, o1, GPT-4o Realtime, gpt-realtime, Qwen3-14B, Qwen3-Reranker, Qwen3-Embedding, DeepSeek R1, Jais 30B, Embeddings 3 Large, DALL·E 3, Llama 3.3 70B, Llama 3 70B, Llama 3.2 90B Vision, Mistral 7B, Mixtral 8x7B, Claude Sonnet 4, Cohere Command R, Cohere Command R+, Cohere Command A, Realtime, Stable Diffusion, K2 Think, Grok 2.5, and Whisper. This enables businesses to achieve competitive advantages and reach strategic business goals.

Specifically, Compass is the ultimate platform for seamless Application Programming Interface (API) management and innovation. For developers, organizations, and visionaries seeking to revolutionize their digital experiences, Compass offers the essential tools to excel in the contemporary interconnected business landscape.

Explore the Models

  • GPT-5: An advanced model with enhanced understanding, reasoning ability, and designed to be more accurate, flexible, and efficient.
  • GPT-5 mini: Offers faster performance, making it ideal for well-defined tasks and clear, focused prompts.
  • GPT-4o: One of the advanced and intelligent models that can perform complex and multi-step tasks.
  • GPT-4o Audio: Experience the future of audio intelligence that revolutionizes audio processing with advanced AI, enhancing transcription, translation, and audio content generation.

  • GPT-4o mini: The most advanced model in the small model's category that supports text inputs and generates text outputs. It is ideal for smaller tasks.

  • GPT-4.1: GPT-4.1 is a multimodal model for complex tasks and is well-suited for problem-solving across domains. It offers increased efficiency for faster processing and a streamlined user experience.

  • o1: The o1 model is designed for complex reasoning. It thinks through multiple steps before responding, allowing it to handle more intricate prompts.

  • o3: A versatile and robust model that excels in various domains. It sets a new benchmark for math, science, coding, and visual reasoning. It also excels in technical writing and instruction-following and can think through multi-step problems involving text, code, and image analysis.

  • o3-mini: The latest compact reasoning model, offering advanced intelligence, including essential developer capabilities such as structured outputs and function calling.

  • GPT Image 1: GPT Image 1 is an image generation and natively multimodal language model that accepts text and image inputs and produces image outputs.

  • GPT-4o Transcribe: A speech-to-text model that utilizes GPT-4 to transcribe audio. 

  • GPT-4o mini Transcribe: A speech-to-text model that uses GPT-4o mini to transcribe audio. 

  • GPT-4o mini TTS: A text-to-speech model built on GPT-4o mini is used to convert text to natural-sounding spoken text. 

  • GPT-4o Realtime: A preview model, capable of responding to audio and text inputs in realtime.

  • DALL·E: DALL·E is an advanced text-to-image generation model that creates high-quality images from text prompts. The model can handle abstract ideas and manipulate perspectives, offering a versatile tool for creative expression.

  • Embeddings 3 Large: One of the most capable embedding models from OpenAI for both English and non-English tasks. The key features of this model include high performance, flexibility to adjust the dimensions using the dimension parameter, and efficiency.

  • Whisper: Converts your audio files to text.

  • Realtime: The Realtime API enables users to build low-latency, multi-modal conversational experiences supporting both text and audio as input and output.

  • Jais 30B: Most advanced, bilingual Arabic-English LLM model, powered by Core42.

  • Llama 3.3 70B: An enhanced version of the Llama 3 70B model. A highly performant, cost-effective model that enables diverse use cases.

  • Llama 3 70B: A highly performant, cost-effective model that enables diverse use cases.

  • Stable Diffusion: A Multimodal Diffusion Transformer (MMDiT) text-to-image model that features greatly improved performance in image quality, typography, complex prompt understanding, and resource efficiency.

  • Mixtral 8x7B: Mixtral is a high-quality sparse mixture of experts model (SMoE) with open weights.

  • Mistral 7B: First dense model, perfect for experimentation, customization, and quick iteration.

  • Claude Sonnet 4: A reasoning model with enhanced performance across multiple domains—especially in coding. It offers cutting-edge capabilities that are well-suited for a wide range of AI applications, from user-facing assistants to high-throughput operational tasks.

  • Cohere Command R: A high-performance language model built for enterprise-scale conversational AI and long-context tasks. It delivers exceptional accuracy, efficiency, and scalability—empowering businesses to deploy advanced capabilities like retrieval-augmented generation with speed and reliability.

  • Cohere Command A
    An advanced model offering enhanced performance across a variety of tasks.

  • gpt-oss-120b
    A highly flexible, open-source model by OpenAI licensed under Apache 2.0. This model is optimised for high-performance tasks. The model is designed for reasoning, tool use, and efficient performance. Compass offers the following model versions:

    • gpt-oss-120b Cerebras: The model is deployed and running on high-performance Cerebras clusters.

    • gpt-oss-120b Core42: The model is deployed and running on the Core42 cloud, located in the UAE region.

    • gpt-oss-120b-core42-amd: The model is deployed and running on high-performance AMD hardware.

    • gpt-oss-120b-qualcomm: The model is deployed and running on high-performance Qualcomm.

  • gpt-oss-20b Core42
    A highly flexible, open-source model by OpenAI licensed under Apache 2.0. This model is ideal for experimentation, customisation, and commercial deployment. The model has the following versions:

    • gpt-oss-20b-core42-amd: The model is deployed and running on high-performance AMD hardware.

    • gpt-oss-20b-qualcomm: The model is deployed and running on high-performance Qualcomm.

  • o4-mini
    Engineered for compact efficiency, it delivers fast, cost-efficient reasoning with exceptional performance in maths, coding, and visual tasks. Ideal for developers and creators seeking speed without compromise.

  • Qwen3-14B
    A 14.8B parameter model, optimised for multilingual reasoning, coding, and agentic tasks. Supports 119 languages, long context (up to 131k tokens), and hybrid reasoning modes for flexible performance.

  • Qwen3-Embedding-8B
    A new series of multilingual embedding models optimised for text ranking, retrieval, classification, and code search. Supports over 100 languages, flexible embedding dimensions (up to 4096), and instruction-aware tuning.

  • Qwen3-Reranker-8B
    An 8B parameter model built to enhance search quality by reranking documents based on query relevance. Supports over 100 languages, understands long texts with a 32k context length, and delivers state-of-the-art performance in text and code retrieval tasks.

  • Cohere Embed 4
    Offers advanced multimodal search across text, images, tables, graphs, code, and diagrams. Supports up to 128K tokens (~200 pages), 100+ languages, and secure deployments (VPC/on-prem) for regulated industries.

  • DeepSeek R1
    Offer enhanced reasoning, math, coding, and accuracy with reduced hallucinations. Adds JSON output, function calling, and improved UI generation.

  • K2 Think: An open-source reasoning model that achieves state-of-the-art performance with 32B parameters. It was developed in the UAE by Mohamed bin Zayed University of Artificial Intelligence (MBZUAI).

    • K2 Think Cerebras: The model is deployed and running on the Cerebras in the US region.

    • K2 Think Core42: The model is deployed and running on the Core42 cloud, located in the UAE region.

  • gpt-realtime: The first general-availability realtime model, capable of responding to audio and text inputs in realtime.

  • Llama 3.2 90B Vision: The Llama 3.2 90B Vision is a collection of pretrained and instruction-tuned image reasoning generative models.

  • Cohere Command R+: A large language model optimized for conversational interaction and long-context tasks.

  • Grok 2.5: A large-scale model developed by xAI that uses a Mixture-of-Experts (MoE) architecture.

  • Batch (Add-on): The Batch Inference feature allows you to process large volumes of requests asynchronously, which do not require immediate response. It is currently supported with the GPT-4o and GPT-4o mini models.

  • Web Search (Add-on): Web Search enables users to retrieve search engine results based on user queries in a single API call. The search results include the frequently clicked links on the destination website. The Web Search API is offered as an Add-on and is currently available for GPT-4.1, GPT-4o, and GPT-4o mini models. The Web Search is enabled along with the supported model subscription.

Experience the Chat with Compass Chat Enterprise

Built on GPT-4o and Jais 30B, Compass Chat Enterprise is designed to make your interactions smarter, more engaging, and incredibly convenient.

Help Center

Visit the Frequently Asked Questions (FAQs) page for any general queries related to Compass.

For Developers

Refer to the API Reference section to learn more about the available endpoints on our Compass platform.