%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
On the evening of August 19, DeepSeek officially announced that the online model version has been upgraded to V3.1. The most significant improvement is that the context length has been extended to 128K, equivalent to handling ultra-long texts of 100,000 to 130,000 Chinese characters, suitable for long document analysis, codebase understanding, and multi-turn dialogue scenarios.
Users can now experience the new version through the official website, App, or WeChat mini program. The API call method remains unchanged, allowing developers to seamlessly switch without any additional adjustments.
This upgrade is not a major version iteration, but an optimization of the V3 model. Testing shows that V3.1 performs 43% better in multi-step reasoning tasks compared to its predecessor, especially in complex tasks such as mathematical calculations, code generation, and scientific analysis, with higher accuracy. At the same time, the occurrence of "hallucinations" (generating false information) has decreased by 38%, significantly enhancing the reliability of the output. In addition, V3.1 has optimized multilingual support, especially improving the processing capability for Asian languages and minority languages.
Although V3.1 brings important improvements, the next-generation large model DeepSeek-R2, which users are more eager for, still does not have an announced release date. Previously, there were market rumors that R2 would be released between August 15 and 30, but insiders close to DeepSeek stated that this information is untrue, and the official currently has no specific release plan. Some foreign media reports suggest that the delay of R2 may be related to hardware issues encountered during training, such as limited chip supply.
DeepSeek's update rhythm indicates that a V4 model may be released before R2. However, the official has always maintained a low profile, emphasizing "we will release it once it is completed," and has not responded to any market speculations.
Experience address: https://chat.deepseek.com/
