Mistral NeMo-Instruct 12B
Mistral NeMo-Instruct 12B is an AI model published by MistralAI, released on 2024-07-18, for 基础大模型, with 120.0B parameters, and 128K tokens context length, requiring about 24.51GB storage, under the Apache 2.0 license.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Mistral NeMo-Instruct-12B是MistralAI开源的最新的120亿参数规模的大语言模型，是其开源的70亿参数规模的Mistral 7B模型的继任者，Mistral NeMo-Instruct-12B是指令微调的版本。这意味着未来MistralAI可能不再会开源Mistral 7B，而是用Mistral Nemo代替。
尽管该模型在MMLU得分上没有很突出，但是在推理等任务的表现上效果很好。
欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送
