OpenAI launches open-weight models GPT OSS 120b & GOT OSS 20b: Check how to get new models on your computer
OpenAI has released its first public AI models since 2019, introducing open-weight reasoning models that balance accessibility with proprietary control. Unlike fully open-source alternatives that reveal source code and training data, these models provide only the trained weights. This strategic compromise allows developers to run and adapt the AI locally or in the cloud while enabling OpenAI to retain security over its core intellectual property and training methodologies. The shift addresses intense market competition from other major tech firms and acknowledges the growing demand for transparent, adaptable AI tools. By integrating these models into Microsoft’s cloud platform and supporting local deployment on consumer hardware, OpenAI lowers the barrier to entry for businesses and individual developers. This move enables users to fine-tune the models for specific needs, offering greater flexibility than traditional black-box API services. This development is significant for open data because it expands the pool of usable AI assets beyond strictly proprietary systems. While not fully open-source, providing access to weights facilitates transparency in how models behave and allows for community-driven improvements through fine-tuning. It marks a crucial step toward a more inclusive AI ecosystem, encouraging innovation through accessible infrastructure while maintaining responsible deployment standards.
Source: economictimes.indiatimes.comPublished on 2025-08-07
Related news
- AWS Offers OpenAI’s Models on Its Platform for the First Time | PYMNTS.com
- OpenAI rediscovers an open AI mission, with new reasoning models
- Today’s Cache | OpenAI releases open-weight reasoning models; Anthropic releases Claude Opus 4.1; Intel struggles with manufacturing next PC chip
- OpenAI Returns To Open-source With First Free Models In Six Years
- The US Makes Its First Modern Foray Into Open-Source Models With GPT-OSS, But How Does It Stack Up Against Chinese Counterparts? Hint: Pretty Close