Mistral 7B
This work addresses the need for more efficient and powerful language models for AI applications, offering a competitive open-source alternative.
The authors introduced Mistral 7B, a 7-billion-parameter language model designed for high performance and efficiency, which outperforms larger models like Llama 2 13B across benchmarks and Llama 1 34B in reasoning, mathematics, and code generation.
We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle sequences of arbitrary length with a reduced inference cost. We also provide a model fine-tuned to follow instructions, Mistral 7B -- Instruct, that surpasses the Llama 2 13B -- Chat model both on human and automated benchmarks. Our models are released under the Apache 2.0 license.