Mixtral 8x7B, a high-quality sparse mixture of experts model (SMoE) with open weights. Mixtral outperforms Llama 2 70B on most benchmarks with 6x faster inference. - 智能 AI 工具，助力您的工作效率。
3
Views
0
Likes
Jan 2026
Added
github.com
Website
Mixtral 8x7B, a high-quality sparse mixture of experts model (SMoE) with open weights. Mixtral outperforms Llama 2 70B on most benchmarks with 6x faster inference.
Mixtral 是 open-source-llm 类别中的优秀工具，适合需要 AI 辅助的各类用户。
Visit the official website to get started
Have an AI tool to share?
Submit Your ToolDeepSeek's first-generation reasoning models. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning without supervised fine-tuning, demonstrated remarkable performance on reasoning. - 智能 AI 工具，助力您的工作效率。
A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. - 智能 AI 工具，助力您的工作效率。
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud. - 智能 AI 工具，助力您的工作效率。
Llama3 is a large language model developed by Meta AI. It is the successor to Meta's Llama2 language model. - 智能 AI 工具，助力您的工作效率。