Launch Faster with
Pre-Installed Apps.

Meta's open-source 70B instruction model with strong reasoning and multilingual support.

Vision-language model supporting image understanding, OCR, and visual Q&A at 90B scale.

Mixture-of-experts architecture. 16 experts × 17B parameters for efficient large-scale inference.

DS

Chain-of-thought reasoning model optimised for math, code, and complex problem solving.

Compact R1 variant delivering high reasoning performance with lower resource requirements.

Google's open-weight 27B model optimised for instruction following and long-context tasks.

Open-source GPT-class model at 120B scale for enterprise-grade language tasks and generation.

Lightweight GPT-OSS variant for low-latency inference, suitable for real-time applications.

Compact 4B vision-language model for lightweight multimodal inference.

Balanced 8B vision-language model with strong OCR and image understanding.

Full-size 32B vision-language model for demanding multimodal workloads.

High-accuracy OCR toolkit supporting 80+ languages, layout analysis, and table recognition.

High-quality web-based text-to-speech interface with natural voice synthesis.

Pre-configured Docker with NVIDIA Container Toolkit for GPU-accelerated containers.

Interactive notebook environment for Python, data science, and machine learning workflows.

Text-to-speech engine with multi-voice support and fast real-time synthesis output.

Expressive conversational TTS model with controllable emotion and voice cloning.

OpenAI Whisper with a web interface for accurate speech-to-text transcription and translation.