DeepSeek-R1
DeepSeek-R1 is a large language reasoning model developed by DeepSeek, focused on complex logical reasoning, mathematical computation, and code generation tasks. The model is trained using reinforcement learning and chain-of-thought techniques, exhibiting strong multi-step reasoning capabilities. It supports open-source deployment and can run on a variety of hardware platforms, representing a flagship version of DeepSeek's reasoning model series.
Core Facts
Reasoning models, represented by OpenAI o1, DeepSeek-R1, and QwQ, have chain-of-thought expansion as their core mechanism, generating a large number of intermediate reasoning steps before producing a final answer.
DeepSeek-V3 adopts a Mixture of Experts (MoE) architecture with a total parameter count of 671 billion, but only approximately 37 billion parameters are activated per inference.
According to official disclosure, DeepSeek's training cost was only approximately $6 million.
Timeline (last 90 days)
GPT's o1, DeepSeek-R1, and other reasoning models correspond to System 2 and first generate a chain of thought before answering, at the cost of high latency and large token consumption; decision models correspond to System 1 and map inputs directly to labels.
DeepSeek-R1, Claude Sonnet 5.5, and Claude Opus 5.5 scored 0.9375 on resolve but dropped to 0.875 on resolve-plus; DeepSeek-R1 leads the open-source group with these two scores.
When placing reasoning models behind strict micro-contracts, the thinking channel must be separated or stripped; otherwise the model will fail on protocol syntax.
DeepSeek-R1 scored zero across all 120 trials because it outputs raw <think> tokens into the response stream, exceeding the 256-token output budget before producing valid JSON.
DeepSeek-R1 repeatedly produces "wait" during reasoning, manifesting as linear enumeration of possibilities accompanied by repeated self-doubt.
Open-source models such as QwQ and DeepSeek-R1 are already able to compete on par with top closed-source commercial models on this navigation task.
The reasoning model DeepSeek-R1 scored only 0.138, with most answers missing required fields or being unparseable, because the budget was spent on thinking rather than outputting JSON.
gpt-oss-120b scored 0.991 on memory read but only 0.740 on write, ranking last in write; DeepSeek-R1 scored 0.997 on read and 0.863 on write.
Larger, newer models do not necessarily perform better: gpt-oss-120b (0.740) scored lower than gpt-oss-20b (0.761); DeepSeek-R1, costing $1.94, scored lower than Gemma 4 31B, which cost only 13 cents.
The deployment environment demonstrated in the tutorial includes Windows 11, Docker, Ollama + DeepSeek-R1, and Dify open-source community edition 0.15.3, with minimum hardware requirements of CPU ≥ 2 cores and memory ≥ 4 GB.
39 more timeline events
All Facts (20)
Reasoning models, represented by OpenAI o1, DeepSeek-R1, and QwQ, have chain-of-thought expansion as their core mechanism, generating a large number of intermediate reasoning steps before producing a final answer.
90%VerifiedDeepSeek-V3 adopts a Mixture of Experts (MoE) architecture with a total parameter count of 671 billion, but only approximately 37 billion parameters are activated per inference.
90%VerifiedAccording to official disclosure, DeepSeek's training cost was only approximately $6 million.
80%VerifiedDeepSeek-R1 is a model focused on reasoning capabilities and possesses chain-of-thought ability through reinforcement learning training.
80%VerifiedThe OpenAI o1 series models were released in September 2024, representing a paradigm shift in LLMs from fast, intuitive responses to deep reasoning.
80%VerifiedDeepSeek R1 offers multiple distilled versions ranging from 1.5B to 671B parameters.
80%VerifiedOpenAI的o1系列模型和DeepSeek-R1采用了类似Chain-of-Thought的显式推理机制
80%VerifiedIn early 2025, DeepSeek-R1 achieved reasoning capabilities comparable to OpenAI o1 at a training cost of approximately $6 million.
80%VerifiedOpenAI's o1/o3 series and Google's Gemini Thinking adopt a slow-thinking mechanism similar to extended thinking.
80%VerifiedThe DeepSeek-R1 technical report indicates that this training approach can cause the model to spontaneously develop complex reasoning behaviors such as self-correction and re-thinking, without requiring explicit human annotation guidance
75%VerifiedBefore outputting a final answer, a reasoning model's internal chain of thought may consume 3,000 to 20,000 tokens, far exceeding the approximately 500 tokens of an ordinary model.
75%VerifiedOpenAI o1 and DeepSeek-R1 achieve breakthroughs on complex benchmarks such as math competition problems by extending the chain of thought, which is essentially dynamically allocating more computational budget during the inference stage.
75%VerifiedDeepSeek-R1通过在代码和数学两个可验证域上大规模运行GRPO算法取得成功
70%VerifiedMainstream open-source models such as Meta's Llama series, DeepSeek-R1, and Mistral mostly fall under the category of open-weight models.
70%VerifiedDeepMind research has shown that on certain mathematical reasoning tasks, smaller models can surpass models with larger parameter counts by increasing the amount of inference-time computation.
70%VerifiedGRPO is a reinforcement learning algorithm proposed by the DeepSeek team that eliminates the Critic Model from PPO, computing intra-group relative rewards as a baseline by generating a set of responses to the same question.
65%VerifiedDeepSeek-R1 and the Qwen series have achieved comparable reasoning capabilities to GPT-4 at a training cost far lower than GPT-4's.
65%VerifiedDeepSeek achieved mathematical and coding capabilities close to top closed-source models at extremely low training cost, and its R1 model attracted global attention upon release.
65%VerifiedReasoning models, represented by OpenAI's o series, DeepSeek-R1, and the Kimi K series, perform a chain-of-thought process before outputting, decomposing problems into multiple steps, performing self-verification, and deriving answers incrementally.
65%VerifiedGoogle's Gemini 2.0 Flash Thinking and models such as DeepSeek-R1 have followed the test-time compute scaling technical path of o1.
65%