The team introduces its first - generation reasoning models, DeepSeek - R1 - Zero and DeepSeek - R1. DeepSeek - R1 - Zero, trained via large - scale RL without SFT, shows remarkable reasoning ability but also has problems like endless repetition. DeepSeek - R1, which incorporates cold - start data before RL, solves these issues and achieves performance on par with OpenAI - o1. The team has open - sourced these models and six distilled ones, with DeepSeek - R1 - Distill - Qwen - 32B outperforming OpenAI - o1 - mini in benchmarks.
Discover more sites in the same category
The open source AI model you can fine-tune, distill and deploy anywhere. Our latest models are available in 8B, 70B, and 405B variants.
The open source AI model you can fine-tune, distill and deploy anywhere. Our latest models are available in 8B, 70B, and 405B variants.
阿里巴巴开源Qwen3大模型,2350亿参数支持119种语言,首创"快思考/慢思考"混合推理,数学/代码能力超越Gemini 2.5 Pro,4张显卡即可部署
WanAI is an AI-powered creative drawing tool that leverages advanced artificial intelligence and large-scale models to generate artwork. It enables users to create unique paintings and illustrations by inputting prompts or selecting from various styles. The platform is designed to cater to both amateur and professional artists, providing an intuitive interface and a wide range of customization options. With WanAI, users can explore new creative possibilities and streamline their artistic workflows.
Share your thoughts about this page. All fields marked with * are required.