DeepSeek-R1-0120 API, Deployment, Pricing

DeepSeek-R1-0528 is a reasoning model powered by reinforcement learning (RL) that addresses the issues of repetition and readability. Prior to RL, DeepSeek-R1 incorporated cold-start data to further optimize its reasoning performance. It achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks, and through carefully designed training methods, it has enhanced overall effectiveness...

DeepSeek-R1-Distill-Llama-70B is a distilled model based on Llama-3.3-70B-Instruct. As part of the DeepSeek-R1 series, it was fine-tuned using samples generated by DeepSeek-R1 and demonstrates excellent performance across mathematics, programming, and reasoning tasks. The model achieved impressive results in various benchmarks including AIME 2024, MATH-500, and GPQA Diamond, showcasing its strong reasoning capabilities...

DeepSeek-R1-Distill-Llama-8B is a distilled model based on Llama-3.1-8B. The model was fine-tuned using samples generated by DeepSeek-R1 and demonstrates strong reasoning capabilities. It achieved notable results across various benchmarks, including 89.1% accuracy on MATH-500, 50.4% pass rate on AIME 2024, and a rating of 1205 on CodeForces, showing impressive mathematical and programming abilities for an 8B-scale model...

DeepSeek-R1-Distill-Qwen-1.5B is a distilled model based on Qwen2.5-Math-1.5B. The model was fine-tuned using 800k curated samples generated by DeepSeek-R1 and demonstrates decent performance across various benchmarks. As a lightweight model, it achieved 83.9% accuracy on MATH-500, 28.9% pass rate on AIME 2024, and a rating of 954 on CodeForces, showing reasoning capabilities beyond its parameter scale...

DeepSeek-R1-Distill-Qwen-14B is a distilled model based on Qwen2.5-14B. The model was fine-tuned using 800k curated samples generated by DeepSeek-R1 and demonstrates strong reasoning capabilities. It achieved impressive results across various benchmarks, including 93.9% accuracy on MATH-500, 69.7% pass rate on AIME 2024, and a rating of 1481 on CodeForces, showcasing its powerful abilities in mathematics and programming tasks...

DeepSeek-R1-Distill-Qwen-32B is a distilled model based on Qwen2.5-32B. The model was fine-tuned using 800k curated samples generated by DeepSeek-R1 and demonstrates exceptional performance across mathematics, programming, and reasoning tasks. It achieved impressive results in various benchmarks including AIME 2024, MATH-500, and GPQA Diamond, with a notable 94.3% accuracy on MATH-500, showcasing its strong mathematical reasoning capabilities...

DeepSeek-R1-Distill-Qwen-7B is a distilled model based on Qwen2.5-Math-7B. The model was fine-tuned using 800k curated samples generated by DeepSeek-R1 and demonstrates strong reasoning capabilities. It achieved impressive results across various benchmarks, including 92.8% accuracy on MATH-500, 55.5% pass rate on AIME 2024, and a rating of 1189 on CodeForces, showing remarkable mathematical and programming abilities for a 7B-scale model...

The new version of DeepSeek-V3 (DeepSeek-V3-0324) utilizes the same base model as the previous DeepSeek-V3-1226, with improvements made only to the post-training methods. The new V3 model incorporates reinforcement learning techniques from the training process of the DeepSeek-R1 model, significantly enhancing its performance on reasoning tasks. It has achieved scores surpassing GPT-4.5 on evaluation sets related to mathematics and coding. Additionally, the model has seen notable improvements in tool invocation, role-playing, and casual conversation capabilities....

DeepSeek-V3.1 is a hybrid large language model released by DeepSeek AI, featuring significant upgrades over its predecessor. A key innovation is the integration of both a 'Thinking Mode' for deliberative, chain-of-thought reasoning and a 'Non-thinking Mode' for direct responses, which can be switched via the chat template to suit various tasks. The model's capabilities in tool use and agent tasks have been substantially improved through post-training optimization, enabling better support for external search tools and complex multi-step instructions. DeepSeek-V3.1 is post-trained on top of the DeepSeek-V3.1-Base model, which underwent a two-phase long-context extension with a vastly expanded dataset, enhancing its ability to process long documents and codebases. As an open-source model, DeepSeek-V3.1 demonstrates performance comparable to leading closed-source models on various benchmarks, particularly in coding, math, and reasoning, while its Mixture-of-Experts (MoE) architecture maintains a massive parameter count while reducing inference costs...

DeepSeek-VL2 is a mixed-expert (MoE) vision-language model developed based on DeepSeekMoE-27B, employing a sparse-activated MoE architecture to achieve superior performance with only 4.5B active parameters. The model excels in various tasks including visual question answering, optical character recognition, document/table/chart understanding, and visual grounding. Compared to existing open-source dense models and MoE-based models, it demonstrates competitive or state-of-the-art performance using the same or fewer active parameters....

ERNIE-4.5-300B-A47B is a large language model developed by Baidu based on a Mixture-of-Experts (MoE) architecture. The model has a total of 300 billion parameters, but only activates 47 billion parameters per token during inference, thus balancing powerful performance with computational efficiency. As one of the core models in the ERNIE 4.5 series, it is trained on the PaddlePaddle deep learning framework and demonstrates outstanding capabilities in tasks such as text understanding, generation, reasoning, and coding. The model utilizes an innovative multimodal heterogeneous MoE pre-training method, which effectively enhances its overall abilities through joint training on text and visual modalities, showing prominent results in instruction following and world knowledge memorization. Baidu has open-sourced this model along with others in the series to promote the research and application of AI technology...

GLM-4-32B-0414 is a new generation model in the GLM family with 32 billion parameters. Its performance is comparable to OpenAI's GPT series and DeepSeek's V3/R1 series, and it supports very user-friendly local deployment features. GLM-4-32B-Base-0414 was pre-trained on 15T of high-quality data, including a large amount of reasoning-type synthetic data, laying the foundation for subsequent reinforcement learning extensions. In the post-training stage, in addition to human preference alignment for dialogue scenarios, the team enhanced the model's performance in instruction following, engineering code, and function calling using techniques such as rejection sampling and reinforcement learning, strengthening the atomic capabilities required for agent tasks. GLM-4-32B-0414 achieves good results in areas such as engineering code, Artifact generation, function calling, search-based Q&A, and report generation. On several benchmarks, its performance approaches or even exceeds that of larger models like GPT-4o and DeepSeek-V3-0324 (671B)...