GitHub – Deepseek-ai/DeepSeek-V3

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language design with 671B overall specifications with 37B triggered for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were completely verified in DeepSeek-V2. Furthermore, DeepSeek-V3 leaders an auxiliary-loss-free method for load balancing and sets a multi-token prediction training objective for more powerful efficiency. We pre-train DeepSeek-V3 on 14.8 trillion varied and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning phases to fully harness its abilities. Comprehensive assessments reveal that DeepSeek-V3 exceeds other open-source models and attains performance comparable to leading closed-source designs. Despite its outstanding performance, DeepSeek-V3 needs only 2.788 M H800 GPU hours for its full training. In addition, its training process is incredibly stable. Throughout the whole training procedure, we did not experience any irrecoverable loss spikes or carry out any rollbacks.

2. Model Summary

Architecture: Innovative Load Balancing Strategy and Training Objective

– On top of the effective architecture of DeepSeek-V2, we pioneer an auxiliary-loss-free method for load balancing, which reduces the efficiency deterioration that emerges from motivating load balancing.
– We examine a Multi-Token Prediction (MTP) goal and show it beneficial to model performance. It can also be utilized for speculative decoding for reasoning acceleration.

Pre-Training: Towards Ultimate Training Efficiency

– We create an FP8 combined precision training framework and, for the very first time, confirm the expediency and effectiveness of FP8 training on an extremely large-scale design.
– Through co-design of algorithms, structures, and hardware, we overcome the communication bottleneck in cross-node MoE training, almost accomplishing full computation-communication overlap.
This significantly enhances our training performance and minimizes the training expenses, enabling us to even more scale up the design size without additional overhead.
– At a cost-effective cost of just 2.664 M H800 GPU hours, we finish the pre-training of DeepSeek-V3 on 14.8 T tokens, producing the currently greatest open-source base design. The subsequent training stages after pre-training require only 0.1 M GPU hours.

Post-Training: Knowledge Distillation from DeepSeek-R1

– We present an ingenious method to boil down thinking capabilities from the long-Chain-of-Thought (CoT) model, specifically from one of the DeepSeek R1 series designs, into standard LLMs, especially DeepSeek-V3. Our pipeline elegantly incorporates the verification and reflection patterns of R1 into DeepSeek-V3 and especially enhances its thinking efficiency. Meanwhile, we likewise maintain a control over the output style and length of DeepSeek-V3.

3. Model Downloads

The overall size of DeepSeek-V3 designs on Hugging Face is 685B, which consists of 671B of the Main Model weights and 14B of the Multi-Token Prediction (MTP) Module weights. **

To ensure ideal performance and flexibility, we have actually partnered with open-source neighborhoods and hardware vendors to offer multiple ways to run the design in your area. For detailed assistance, have a look at Section 6: How_to Run_Locally.

For designers looking to dive much deeper, we recommend checking out README_WEIGHTS. md for information on the Main Model weights and the Multi-Token Prediction (MTP) Modules. Please note that MTP assistance is presently under active development within the neighborhood, and we welcome your contributions and feedback.

4. Evaluation Results

Base Model

Standard Benchmarks

Best outcomes are displayed in bold. Scores with a gap not surpassing 0.3 are considered to be at the exact same level. DeepSeek-V3 accomplishes the best efficiency on many criteria, particularly on math and code jobs. For more examination information, please check our paper.

Context Window

Evaluation results on the Needle In A Haystack (NIAH) tests. DeepSeek-V3 carries out well throughout all context window lengths as much as 128K.

Chat Model

Standard Benchmarks (Models larger than 67B)

All designs are examined in a setup that restricts the output length to 8K. Benchmarks consisting of fewer than 1000 samples are checked numerous times using differing temperature level settings to obtain robust final outcomes. DeepSeek-V3 stands as the best-performing open-source model, and likewise displays competitive efficiency versus frontier closed-source models.

Open Ended Generation Evaluation

English open-ended discussion evaluations. For AlpacaEval 2.0, we use the length-controlled win rate as the metric.

5. Chat Website & API Platform

You can chat with DeepSeek-V3 on DeepSeek’s official site: chat.deepseek.com

We likewise offer OpenAI-Compatible API at DeepSeek Platform: platform.deepseek.com

6. How to Run Locally

DeepSeek-V3 can be deployed locally using the following hardware and open-source neighborhood software application:

DeepSeek-Infer Demo: We supply a simple and light-weight demonstration for FP8 and BF16 reasoning.
SGLang: Fully support the DeepSeek-V3 design in both BF16 and FP8 reasoning modes, with Multi-Token Prediction coming quickly.
LMDeploy: Enables efficient FP8 and BF16 reasoning for local and cloud deployment.
TensorRT-LLM: Currently supports BF16 inference and INT4/8 quantization, with FP8 support coming quickly.
vLLM: Support DeepSeek-V3 design with FP8 and BF16 modes for tensor parallelism and pipeline parallelism.
AMD GPU: Enables running the DeepSeek-V3 model on AMD GPUs through SGLang in both BF16 and FP8 modes.
Huawei Ascend NPU: Supports running DeepSeek-V3 on Huawei Ascend devices.
Since FP8 training is natively embraced in our structure, we only offer FP8 weights. If you need BF16 weights for experimentation, you can use the offered conversion script to perform the change.

Here is an example of converting FP8 weights to BF16:

Hugging Face’s Transformers has actually not been straight supported yet. **

6.1 Inference with DeepSeek-Infer Demo (example only)

System Requirements

Note

Linux with Python 3.10 only. Mac and Windows are not supported.

Dependencies:

Model Weights & Demo Code Preparation

First, clone our DeepSeek-V3 GitHub repository:

Navigate to the inference folder and install reliances noted in requirements.txt. Easiest way is to utilize a package supervisor like conda or uv to create a brand-new virtual environment and set up the dependences.

Download the design weights from Hugging Face, and put them into/ path/to/DeepSeek-V 3 folder.

Model Weights Conversion

Convert Hugging Face model weights to a specific format:

Run

Then you can talk with DeepSeek-V3:

Or batch reasoning on a provided file:

6.2 Inference with SGLang (advised)

SGLang presently supports MLA optimizations, DP Attention, FP8 (W8A8), FP8 KV Cache, and Torch Compile, delivering modern latency and throughput performance among open-source frameworks.

Notably, SGLang v0.4.1 fully supports running DeepSeek-V3 on both NVIDIA and AMD GPUs, making it an extremely flexible and robust solution.

SGLang also supports multi-node tensor parallelism, allowing you to run this model on multiple network-connected machines.

Multi-Token Prediction (MTP) remains in development, and development can be tracked in the optimization plan.

Here are the launch directions from the SGLang team: https://github.com/sgl-project/sglang/tree/main/benchmark/deepseek_v3

6.3 Inference with LMDeploy (advised)

LMDeploy, a flexible and high-performance inference and serving structure customized for big language models, now supports DeepSeek-V3. It offers both offline pipeline processing and online deployment abilities, effortlessly incorporating with PyTorch-based workflows.

For comprehensive step-by-step instructions on running DeepSeek-V3 with LMDeploy, please refer to here: InternLM/lmdeploy # 2960

6.4 Inference with TRT-LLM (recommended)

TensorRT-LLM now supports the DeepSeek-V3 model, offering accuracy options such as BF16 and INT4/INT8 weight-only. Support for FP8 is currently in development and will be released soon. You can access the custom branch of TRTLLM specifically for DeepSeek-V3 support through the following link to experience the brand-new features straight: https://github.com/NVIDIA/TensorRT-LLM/tree/deepseek/examples/deepseek_v3.

6.5 Inference with vLLM (recommended)

vLLM v0.6.6 supports DeepSeek-V3 inference for FP8 and BF16 modes on both NVIDIA and AMD GPUs. Aside from standard methods, vLLM offers allowing you to run this design on several machines linked by networks. For detailed assistance, please refer to the vLLM instructions. Please feel totally free to follow the improvement strategy as well.

6.6 Recommended Inference Functionality with AMD GPUs

In partnership with the AMD group, we have accomplished Day-One support for AMD GPUs utilizing SGLang, with full compatibility for both FP8 and BF16 accuracy. For comprehensive assistance, please describe the SGLang instructions.

6.7 Recommended Inference Functionality with Huawei Ascend NPUs

The MindIE framework from the Huawei Ascend neighborhood has successfully adapted the BF16 version of DeepSeek-V3. For step-by-step guidance on Ascend NPUs, please follow the guidelines here.

7. License

This code repository is accredited under the MIT License. The usage of DeepSeek-V3 Base/Chat models is subject to the Model License. DeepSeek-V3 series (consisting of Base and Chat) supports business use.