deepseek-ai / DeepSeek-V2

Model Download | Evaluation Results | Model Architecture | API Platform | License | Citation Paper Link 👁️ DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model • Introduction Today, we’re introducing DeepSeek-V2, a strong Mixture-of-Experts (MoE) language model characterized by economical training and efficient inference. It comprises 236B total parameters, of which 21B are activated for each token. Compared with DeepSeek 67B, DeepSeek-V2 achieves stronger performance, and meanwhile saves 42.5% of training costs, reduces the KV cache by 93.3%, and boosts the maximum generation throughput to 5.76 times. We pretrained DeepSeek-V2 on a diverse and high-quality corpus comprising 8.1 trillion tokens. This comprehensive pretraining was followed by a process of Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) to fully unleash the model's capabilities. The evaluation results validate the effectiveness of our approach as DeepSeek-V2 achieves remarkable performance on both standard benchmarks and open-ended generation evaluation. • News • 2024.05.16: We released the DeepSeek-V2-Lite. • 2024.05.06: We released the DeepSeek-V2. • Model Downloads | **Model** | **#Total Params** | **#Activated Params** | **Context Length** | **Download** | | :------------: | :------------: | :------------: | :------------: | :------------: | | DeepSeek-V2-Lite | 16B | 2.4B | 32k | 🤗 HuggingFace | | DeepSeek-V2-Lite-Chat (SFT) | 16B | 2.4B | 32k | 🤗 HuggingFace | | DeepSeek-V2 | 236B | 21B | 128k | 🤗 HuggingFace | | DeepSeek-V2-Chat (RL) | 236B | 21B | 128k | 🤗 HuggingFace | Due to the constraints of HuggingFace, the open-source code currently experiences slower performance than our internal codebase when running on GPUs with Huggingface. To facilitate the efficient execution of our model, we offer a dedicated vllm solution that optimizes performance for running our model effectively. • Evaluation Results Base Model Standard Benchmark (Models larger than 67B) | **Benchmark** | **Domain** | **LLaMA3 70B** | **Mixtral 8x22B** | **DeepSeek-V1 (Dense-67B)** | **DeepSeek-V2 (MoE-236B)** | |:-----------:|:--------:|:------------:|:---------------:|:-------------------------:|:------------------------:| | **MMLU** | English | 78.9 | 77.6 | 71.3 | 78.5 | | **BBH** | English | 81.0 | 78.9 | 68.7 | 78.9 | | **C-Eval** | Chinese | 67.5 | 58.6 | 66.1 | 81.7 | | **CMMLU** | Chinese | 69.3 | 60.0 | 70.8 | 84.0 | | **HumanEval** | Code | 48.2 | 53.1 | 45.1 | 48.8 | | **MBPP** | Code | 68.6 | 64.2 | 57.4 | 66.6 | | **GSM8K** | Math | 83.0 | 80.3 | 63.4 | 79.2 | | **Math** | Math | 42.2 | 42.5 | 18.7 | 43.6 | Standard Benchmark (Models smaller than 16B) | **Benchmark** | **Domain** | **DeepSeek 7B (Dense)** | **DeepSeekMoE 16B** | **DeepSeek-V2-Lite (MoE-16B)** | |:-------------:|:----------:|:--------------:|:-----------------:|:--------------------------:| | **Architecture** | - | MHA+Dense | MHA+MoE | MLA+MoE | | **MMLU** | English | 48.2 | 45.0 | 58.3 | | **BBH** | English | 39.5 | 38.9 | 44.1 | | **C-Eval** | Chinese | 45.0 | 40.6 | 60.3 | | **CMMLU** | Chinese | 47.2 | 42.5 | 64.3 | | **HumanEval** | Code | 26.2 | 26.8 | 29.9 | | **MBPP** | Code | 39.0 | 39.2 | 43.2 | | **GSM8K** | Math | 17.4 | 18.8 | 41.1 | | **Math** | Math | 3.3 | 4.3 | 17.1 | For more evaluation details, such as few-shot settings and prompts, please check our paper. Context Window Evaluation results on the Needle In A Haystack (NIAH) tests. DeepSeek-V2 performs well across all context window lengths up to **128K**. Chat Model Standard Benchmark (Models larger than 67B) | Benchmark | Domain | QWen1.5 72B Chat | Mixtral 8x22B | LLaMA3 70B Instruct | DeepSeek-V1 Chat (SFT) | DeepSeek-V2 Chat (SFT) | DeepSeek-V2 Chat (RL) | |:-----------:|:----------------:|:------------------:|:---------------:|:---------------------:|:-------------:|:-----------------------:|:----------------------:| | **MMLU** | English | 76.2 | 77.8 | 80.3 | 71.1 | 78.4 | 77.8 | | **BBH** | English | 65.9 | 78.4 | 80.1 | 71.7 | 81.3 | 79.7 | | **C-Eval** | Chinese | 82.2 | 60.0 | 67.9 | 65.2 | 80.9 | 78.0 | | **CMMLU** | Chinese | 82.9 | 61.0 | 70.7 | 67.8 | 82.4 | 81.6 | | **HumanEval** | Code | 68.9 | 75.0 | 76.2 | 73.8 | 76.8 | 81.1 | | **MBPP** | Code | 52.2 | 64.4 | 69.8 | 61.4 | 70.4 | 72.0 | | **LiveCodeBench (0901-0401)** | Code | 18.8 | 25.0 | 30.5 | 18.3 | 28.7 | 32.5 | | **GSM8K** | Math | 81.9 | 87.9 | 93.2 | 84.1 | 90.8 | 92.2 | | **Math** | Math | 40.6 | 49.8 | 48.5 | 32.6 | 52.7 | 53.9 | Standard Benchmark (Models smaller than 16B) | Benchmark | Domain | DeepSeek 7B Chat (SFT) | DeepSeekMoE 16B Chat (SFT) | DeepSeek-V2-Lite 16B Chat (SFT) | |:-----------:|:----------------:|:------------------:|:---------------:|:---------------------:| | **MMLU** | English | 49.7 | 47.2 | 55.7 | | **BBH** | English | 43.1 | 42.2 | 48.1 | | **C-Eval** | Chinese | 44.7 | 40.0 | 60.1 | | **CMMLU** | Chinese | 51.2 | 49.3 | 62.5 | | **HumanEval** | Code | 45.1 | 45.7 | 57.3 | | **MBPP** | Code | 39.0 | 46.2 | 45.8 | | **GSM8K** | Math | 62.6 | 62.2 | 72.0 | | **Math** | Math | 14.7 | 15.2 | 27.9 | English Open Ended Generation Evaluation We evaluate our model on AlpacaEval 2.0 and MTBench, showing the competitive performance of DeepSeek-V2-Chat-RL on English conversation generation. Chinese Open Ended Generation Evaluation **Alignbench** (https://arxiv.org/abs/2311.18743) | **模型** | **开源/闭源** | **总分** | **中文推理** | **中文语言** | | :---: | :---: | :---: | :---: | :---: | | gpt-4-1106-preview | 闭源 | 8.01 | 7.73 | 8.29 | | DeepSeek-V2 Chat (RL) | 开源 | 7.91 | 7.45 | 8.36 | | erniebot-4.0-202404 (文心一言) | 闭源 | 7.89 | 7.61 | 8.17 | | DeepSeek-V2 Chat (SFT) | 开源 | 7.74 | 7.30 | 8.17 | | gpt-4-0613 | 闭源 | 7.53 | 7.47 | 7.59 | | erniebot-4.0-202312 (文心一言) | 闭源 | 7.36 | 6.84 | 7.88 | | moonshot-v1-32k-202404 (月之暗面) | 闭源 | 7.22 | 6.42 | 8.02 | | Qwen1.5-72B-Chat (通义千问) | 开源 | 7.19 | 6.45 | 7.93 | | DeepSeek-67B-Chat | 开源 | 6.43 | 5.75 | 7.11…