Falcon Mamba 7B, based on the Space Language Model (SSLM) architecture, is the top-performing open-source SSLM, according to Hugging Face. It offers low memory usage and can generate long text blocks without extra memory. Falcon Mamba 7B surpasses traditional models like Meta’s Llama 3.1 8B and Mistral’s 7B.

very nice. 

release: https://falconllm.tii.ae/
model: https://huggingface.co/tiiuae/falcon-mamba-7b

Impressive open-source SSLM with top performance and low memory usage. Surpasses traditional models. Worth checking out