Fast open-model inference API
Accessing fast inference for popular open-source models (Llama, Mixtral, Code Llama) through an OpenAI-compatible API, making it a drop-in replacement in existing code. It's also used for fine-tuning open models on custom datasets without managing GPU clusters.
Create an account at together.ai and copy your API key from the dashboard. Use the OpenAI Python SDK pointed at Together's base URL (api.together.xyz), or install the together Python package directly. New accounts get free credits to test models immediately.
Be the first to share a Together AI case study and get discovered by clients.
Submit a case studyThought leaders
Follow for insights, tutorials, and thought leadership
Together AI
Co-Founder and CEO of Together AI, providing fast and affordable infrastructure for open-source AI models. Building the platform to make open-source AI competitive with closed-source alternatives.
Stanford University / Together AI
Associate Professor of Computer Science at Stanford and Co-Founder of Together AI. Directs Stanford's Center for Research on Foundation Models (CRFM). Created HELM (Holistic Evaluation of Language Models). Pioneering researcher in NLP and AI transparency.
Submit a brief and we'll match you with vetted specialists who have proven Together AI experience.