# Open-source Framework

Eko is a production-grade intelligent agent framework designed for developers. It allows them to easily construct agent-based workflows through natural language and coding logic. Eko's key advantages include efficient task decomposition capabilities, robust tool support, and flexible customization options. It aims to help developers quickly implement complex automated tasks, thereby enhancing development efficiency. Developed by the FellouAI team, Eko is currently open-source and supports various platforms, including browsers and desktop environments. The exact pricing has not been explicitly disclosed, but given its open-source nature, it may be available for free to developers, although some advanced features or customization services might require payment.

FastVideo is an open-source framework designed to accelerate large video diffusion models. It offers two consistency distillation video diffusion models, FastHunyuan and FastMochi, achieving an 8x increase in inference speed. FastVideo introduces the first open video DiT distillation recipe based on PCM (Phased-Consistency-Model), supporting distillation, fine-tuning, and inference for state-of-the-art open video DiT models, including Mochi and Hunyuan. Additionally, FastVideo supports scalable training using FSDP, sequence parallelism, and selective activation checkpoints, as well as memory-efficient fine-tuning with LoRA, pre-computed latent variables, and pre-computed text embeddings. Ongoing development is highly experimental, with future plans to introduce more distillation methods, support additional models, and update the code.