After supervised finetuning the Qwen2. 5-32B-Instruct language model on s1K and equipping it with budget forcing, our model s1-32B exceeds o1-preview on competition math questions by up to 27% (MATH and AIME24).
Ranked #4 on
Mathematical Reasoning
on AIME24
We present DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL, through two key major upgrades.
Ranked #1 on
Referring Expression Comprehension
on RefCOCOg-test
Since the release of ChatGPT, large language models (LLMs) have demonstrated remarkable capabilities across various domains.
In this work, we make the first attempt to fine-tune all-modality models (i. e. input and output with any modality, also named any-to-any models) using human preference data across all modalities (including text, image, audio, and video), ensuring its behavior aligns with human intentions.
In the second stage, we deploy the policies in the real world and collect real-world data to train a delta (residual) action model that compensates for the dynamics mismatch.
We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1.
Ranked #1 on
Mathematical Reasoning
on AIME24
We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.
To create rich visualizations, data analysts often need to iterate back and forth among data processing and chart specification to achieve their goals.
We introduce RAG Foundry, an open-source framework for augmenting large language models for RAG use cases.
The rapid development of large language models has revolutionized code intelligence in software development.
Ranked #4 on
Code Generation
on APPS