DeepSeek Open-Sources DeepSeek-R1 LLM with Performance Comparable To OpenAI's O1 Model

DeepSeek open-sourced DeepSeek-R1, an LLM fine-tuned with reinforcement learning (RL) to improve reasoning capability. DeepSeek-R1 attains results on par with OpenAI's o1 model on numerous criteria, including MATH-500 and SWE-bench.

DeepSeek-R1 is based upon DeepSeek-V3, a mix of experts (MoE) model just recently open-sourced by DeepSeek. This base model is fine-tuned utilizing Group Relative Policy Optimization (GRPO), a reasoning-oriented version of RL. The research study team also performed knowledge distillation from DeepSeek-R1 to open-source Qwen and Llama designs and launched a number of versions of each; these designs outperform larger designs, including GPT-4, on mathematics and coding criteria.

[DeepSeek-R1 is] the primary step toward enhancing language design thinking capabilities utilizing pure reinforcement learning (RL). Our objective is to explore the capacity of LLMs to establish thinking abilities with no supervised information, focusing on their self-evolution through a pure RL process...DeepSeek-R1 ... excels in a broad variety of tasks, including creative writing, basic question answering, modifying, summarization, and more. Additionally, DeepSeek-R1 demonstrates outstanding efficiency on jobs requiring long-context understanding, significantly outshining DeepSeek-V3 on long-context benchmarks.

To develop the model, DeepSeek began with DeepSeek-V3 as a base. They first tried fine-tuning it just with RL, and without any supervised fine-tuning (SFT), producing a model called DeepSeek-R1-Zero, which they have likewise launched. This model displays strong reasoning efficiency, however" powerful reasoning behaviors, it faces numerous problems. For circumstances, DeepSeek-R1-Zero deals with obstacles like poor readability and language blending."

To address this, the group utilized a short stage of SFT to avoid the "cold start" issue of RL. They collected numerous thousand examples of chain-of-thought reasoning to use in SFT of DeepSeek-V3 before running RL. After the RL procedure converged, they then gathered more SFT data utilizing rejection tasting, leading to a dataset of 800k samples. This dataset was used for yewiki.org more fine-tuning and to produce the distilled models from Llama and Qwen.

DeepSeek evaluated their design on a range of thinking, mathematics, and coding criteria and compared it to other models, including Claude-3.5- Sonnet, GPT-4o, and o1. DeepSeek-R1 exceeded all of them on numerous of the benchmarks, consisting of AIME 2024 and MATH-500.

DeepSeek-R1 Performance. Image Source: DeepSeek-R1 Technical Report

Within a few days of its release, the LMArena announced that DeepSeek-R1 was ranked # 3 general in the arena and # 1 in coding and mathematics. It was likewise tied for # 1 with o1 in "Hard Prompt with Style Control" category.

Django framework co-creator Simon Willison blogged about his try outs one of the DeepSeek distilled Llama models on his blog site:

Each response begins with a ... pseudo-XML tag containing the chain of thought used to help create the reaction. [Given the timely] "a joke about a pelican and a walrus who run a tea space together" ... It then believed for 20 paragraphs before outputting the joke! ... [T] he joke is horrible. But the procedure of arriving was such a fascinating insight into how these brand-new designs work.

Andrew Ng's newsletter The Batch blogged about DeepSeek-R1:

DeepSeek is quickly becoming a strong contractor of open designs. Not only are these designs fantastic entertainers, however their license permits use of their outputs for distillation, possibly pressing forward the state of the art for language designs (and multimodal models) of all sizes.

The DeepSeek-R1 models are available on HuggingFace.

About the Author

Anthony Alford

Rate this Article

This content remains in the AI, larsaluarna.se ML & Data Engineering subject