diff --git a/README.md b/README.md index 7ecf87e..c23928d 100644 --- a/README.md +++ b/README.md @@ -47,14 +47,14 @@ ## 1. Introduction -We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. -To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. -Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. +We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters, 37B of which are activated for each token. +To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-Head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. +Furthermore, DeepSeek-V3 introduces an auxiliary loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. -Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training. +Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours to complete training. In addition, its training process is remarkably stable. -Throughout the entire training process, we did not experience any irrecoverable loss spikes or perform any rollbacks. +Throughout the entire training process, we did not experience any irreversible loss spikes or perform any rollbacks. <p align="center"> <img width="80%" src="figures/benchmark.png"> </p> @@ -65,7 +65,7 @@ Throughout the entire training process, we did not experience any irrecoverable **Architecture: Innovative Load Balancing Strategy and Training Objective** -- On top of the efficient architecture of DeepSeek-V2, we pioneer an auxiliary-loss-free strategy for load balancing, which minimizes the performance degradation that arises from encouraging load balancing. +- On top of the efficient architecture of DeepSeek-V2, we pioneer an auxiliary loss-free strategy for load balancing, which minimizes the performance degradation that arises from ensuring load balancing. - We investigate a Multi-Token Prediction (MTP) objective and prove it beneficial to model performance. It can also be used for speculative decoding for inference acceleration. @@ -80,9 +80,9 @@ Throughout the entire training process, we did not experience any irrecoverable --- -**Post-Training: Knowledge Distillation from DeepSeek-R1** +**Post-Training: Knowledge Distilling from DeepSeek-R1** -- We introduce an innovative methodology to distill reasoning capabilities from the long-Chain-of-Thought (CoT) model, specifically from one of the DeepSeek R1 series models, into standard LLMs, particularly DeepSeek-V3. Our pipeline elegantly incorporates the verification and reflection patterns of R1 into DeepSeek-V3 and notably improves its reasoning performance. Meanwhile, we also maintain a control over the output style and length of DeepSeek-V3. +- We introduce an innovative methodology to distill reasoning capabilities from the long-Chain-of-Thought (CoT) model, specifically from one of the DeepSeek R1 series models, into standard LLMs, particularly DeepSeek-V3. Our pipeline elegantly incorporates the verification and reflection patterns of R1 into DeepSeek-V3 and significantly improves its reasoning performance. Meanwhile, we also maintain a control over the output style and length of DeepSeek-V3. --- @@ -99,7 +99,8 @@ Throughout the entire training process, we did not experience any irrecoverable </div> > [!NOTE] -> The total size of DeepSeek-V3 models on Hugging Face is 685B, which includes 671B of the Main Model weights and 14B of the Multi-Token Prediction (MTP) Module weights.** +> The total size of DeepSeek-V3 models on Hugging Face is 685B, which includes 671B of the Main Model weights and 14B of the Multi-Token Prediction (MTP) Module weights +> ** To ensure optimal performance and flexibility, we have partnered with open-source communities and hardware vendors to provide multiple ways to run the model locally. For step-by-step guidance, check out Section 6: [How_to Run_Locally](#6-how-to-run-locally).