commit
62fa6bf18a
1 changed files with 2 additions and 0 deletions
@ -0,0 +1,2 @@ |
|||||
|
<br>DeepSeek open-sourced DeepSeek-R1, an LLM [fine-tuned](https://code.agileum.com) with support knowing (RL) to enhance reasoning capability. DeepSeek-R1 [attains](http://www.boutique.maxisujets.net) results on par with OpenAI's o1 design on several criteria, [including](http://58.34.54.469092) MATH-500 and SWE-bench.<br> |
||||
|
<br>DeepSeek-R1 is based upon DeepSeek-V3, a mix of specialists (MoE) model recently open-sourced by [DeepSeek](https://git.7vbc.com). This [base model](http://git.chuangxin1.com) is [fine-tuned](https://hektips.com) using Group [Relative Policy](https://kanjob.de) [Optimization](https://xn--pm2b0fr21aooo.com) (GRPO), a [reasoning-oriented variant](https://jobs.assist-staffing.com) of RL. The research group likewise performed understanding distillation from DeepSeek-R1 to open-source Qwen and Llama designs and released several [versions](http://51.15.222.43) of each |
Loading…
Reference in new issue