DeepSeek Faces Competition from New AI Models
- DeepSeek caused billions in losses on Wall Street, raising efficiency concerns.
- The Allen Institute released Tülu 3, a 405-billion parameter open-source model.
- Tülu 3 uses Reinforcement Learning with Verifiable Rewards for math tasks.
- Alibaba launched Qwen 2.5-Max, outperforming DeepSeek V3 in key benchmarks.
- Qwen Chat offers text, code, and image generation with web search capabilities.
The release of Tülu 3 and Qwen 2.5-Max challenges DeepSeek’s dominance by introducing models with advanced capabilities in specific areas such as mathematical reasoning and creative writing.
Source (2.6)https://decrypt.co/303703/remember-deepseek-two-new-ai-models-say-theyre-even-better?rand=52368
