https://github.com/SkyworkAI/Skywork-R1V
Pioneering Multimodal Reasoning with CoT
https://github.com/SkyworkAI/Skywork-R1V
deepseek-r1 llm mllm
Last synced: 2 months ago
JSON representation
Pioneering Multimodal Reasoning with CoT
- Host: GitHub
- URL: https://github.com/SkyworkAI/Skywork-R1V
- Owner: SkyworkAI
- License: mit
- Created: 2025-03-15T08:11:44.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-03-26T06:23:17.000Z (3 months ago)
- Last Synced: 2025-03-26T07:26:26.142Z (3 months ago)
- Topics: deepseek-r1, llm, mllm
- Language: Python
- Homepage:
- Size: 32.3 MB
- Stars: 959
- Watchers: 42
- Forks: 82
- Open Issues: 10
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-deep-reasoning - 2025.03.18 - Pioneering Multimodal Reasoning with CoT (News)
- StarryDivineSky - SkyworkAI/Skywork-R1V - R1V 是一个开创性的多模态推理模型,它结合了视觉和语言能力,支持图像理解和推理。该模型的核心特色是 CoT(Chain-of-Thought,思维链)推理,允许模型逐步推导答案,提高准确性和可解释性。Skywork-R1V 在多个基准测试中表现出色,展示了其强大的多模态理解能力。项目提供模型权重、推理代码和训练细节,方便研究人员和开发者使用。它支持多种任务,例如视觉问答、图像描述等。该项目旨在推动多模态人工智能的发展,并为更复杂的人工智能应用铺平道路。Skywork-R1V 的 CoT 推理过程使其能够处理需要逐步分析和推理的任务,而不仅仅是简单的图像识别。模型架构和训练方法的详细信息可以在项目中找到。 (多模态大模型 / 资源传输下载)
- awesome-hacking-lists - SkyworkAI/Skywork-R1V - Pioneering Multimodal Reasoning with CoT (Python)