DeepSeek's new AI model appears to be one of the best 'open' challengers yet
DeepSeek, a Chinese AI firm, launched DeepSeek V3, an open-source model with 671 billion parameters, excelling in text tasks and outperforming competitors, though limited by regulatory constraints.
Read original article
DeepSeek, a Chinese AI firm, has launched DeepSeek V3, a powerful open-source AI model that allows developers to download and modify it for various applications, including commercial use. Released under a permissive license, DeepSeek V3 excels in text-based tasks such as coding, translation, and writing. Internal benchmarks indicate that it outperforms both open and closed AI models, including Meta's Llama 3.1 and OpenAI's GPT-4o, particularly in coding competitions. The model boasts 671 billion parameters and was trained on a dataset of 14.8 trillion tokens, making it significantly larger than its competitors. Despite its capabilities, DeepSeek V3 requires high-end hardware to operate efficiently. The model was developed using a relatively modest budget of $5.5 million and a data center of Nvidia H800 GPUs, which is noteworthy given the restrictions on GPU procurement for Chinese companies. However, the model's responses are influenced by Chinese regulatory requirements, limiting its ability to discuss sensitive topics. DeepSeek is backed by High-Flyer Capital Management, which aims to develop superintelligent AI. The launch of DeepSeek V3 highlights the competitive landscape of AI development, particularly in the open-source domain.
- DeepSeek V3 is one of the most powerful open-source AI models available.
- It outperforms major competitors in various benchmarks, including coding tasks.
- The model has 671 billion parameters and was trained on 14.8 trillion tokens.
- Development costs were significantly lower than those of other leading AI models.
- Regulatory constraints affect the model's ability to discuss sensitive political topics.
Related
Chinese developers scramble as OpenAI blocks access in China
Chinese developers face challenges as OpenAI blocks access in China. SenseTime launches SenseNova 5.5 model to rival GPT-4o. Chinese companies offer free tokens and migration services in response. Tensions between US and China impact AI industry.
DeepSeek v2.5 – open-source LLM comparable to GPT-4o, but 95% less expensive
DeepSeek launched DeepSeek-V2.5, an advanced open-source model with a 128K context length, excelling in math and coding tasks, and offering competitive API pricing for developers.
DeepSeek v3 beats Claude sonnet 3.5 and way cheaper
DeepSeek-V3 is a 671 billion parameter language model that excels in benchmarks, particularly math and coding tasks, utilizing advanced training strategies and supporting various hardware for local deployment.
DeepSeek-V3, ultra-large open-source AI, outperforms Llama and Qwen on launch
Chinese AI startup DeepSeek launched DeepSeek-V3, a 671 billion parameter model outperforming major competitors. It features cost-effective training, innovative architecture, and is available for testing and commercial use.
DeepSeek-V3
DeepSeek has launched DeepSeek-V3, which processes 60 tokens per second, features 671 billion parameters, and maintains open-source compatibility. Pricing changes will occur after February 8, 2024, with future updates planned.
1 commentsRelated
Chinese developers scramble as OpenAI blocks access in China
Chinese developers face challenges as OpenAI blocks access in China. SenseTime launches SenseNova 5.5 model to rival GPT-4o. Chinese companies offer free tokens and migration services in response. Tensions between US and China impact AI industry.
DeepSeek v2.5 – open-source LLM comparable to GPT-4o, but 95% less expensive
DeepSeek launched DeepSeek-V2.5, an advanced open-source model with a 128K context length, excelling in math and coding tasks, and offering competitive API pricing for developers.
DeepSeek v3 beats Claude sonnet 3.5 and way cheaper
DeepSeek-V3 is a 671 billion parameter language model that excels in benchmarks, particularly math and coding tasks, utilizing advanced training strategies and supporting various hardware for local deployment.
DeepSeek-V3, ultra-large open-source AI, outperforms Llama and Qwen on launch
Chinese AI startup DeepSeek launched DeepSeek-V3, a 671 billion parameter model outperforming major competitors. It features cost-effective training, innovative architecture, and is available for testing and commercial use.
DeepSeek-V3
DeepSeek has launched DeepSeek-V3, which processes 60 tokens per second, features 671 billion parameters, and maintains open-source compatibility. Pricing changes will occur after February 8, 2024, with future updates planned.