# Low-Resource Efficient Inference

This is a Mixture of Experts (MoE) model based on Qwen3, with a total of 2.4 billion parameters, including four expert models with 0.6 billion parameters each, designed to deliver more accurate results with higher efficiency and lower memory usage.

Falcon-H1 is a high-performance hybrid architecture language model developed by TII, combining the strengths of Transformers and Mamba architectures, supporting English and multilingual tasks.

Sarvam-1 is a 2 billion parameter language model optimized for Indian languages, excelling in 10 Indian languages.

This is a 2B-parameter Portuguese large language model based on the Gemma architecture, specifically optimized for Brazilian Portuguese, supporting instruction-following and text generation tasks.

DiscoLM German 7B v1 is a 7B-parameter German language model based on the Mistral architecture, supporting both German and English, and released under the Apache-2.0 license.