Together AI is a full-stack AI cloud platform offering serverless inference, fine-tuning, and GPU clusters for open-source AI models. Access 200+ models including Llama, DeepSeek, Qwen, and Flux through a unified API with pay-per-token pricing. Features up to 2.75x faster inference, dedicated endpoints, and costs up to 11x lower than GPT-4. Free tier available with $1 credits.

Replicate is a platform for running and sharing machine learning models, including the FLUX model for image generation.