OpenAI plans open-weight language model launch after six years

OpenAI is set to release its first open-weight language model since GPT-2 in 2019, marking a significant shift in its approach to artificial intelligence development.

CEO Sam Altman announced on March 31 that the new model, designed with enhanced reasoning capabilities, will be made publicly available for developers to use, modify, and deploy on their own hardware.

This model represents a compromise between open-source and proprietary frameworks.

Unlike fully open-source models, the open-weight approach allows access to trained parameters but keeps essential development details confidential.

Altman emphasised that OpenAI aims to gather feedback from developers through events in San Francisco, Europe, and the Asia-Pacific region before finalising the release.

“We’ve been thinking about this for a long time, but other priorities took precedence. Now it feels important to do,” Altman shared on X.

He added that OpenAI is conducting extensive preparatory work, acknowledging that developers will likely modify the model post-release.

The announcement comes amid increasing competition in the AI space.

Rivals like DeepSeek and Alibaba have adopted more open strategies, making their models accessible for experimentation and commercialisation.

OpenAI’s decision to return to an open-weight model reflects growing pressure to balance innovation with accessibility.

Altman also noted during a Reddit Q&A that OpenAI had previously been “on the wrong side of history” regarding open-source initiatives.

While not all members of OpenAI share this perspective, discussions about alternative strategies are ongoing.

The upcoming model could reshape industry dynamics by offering developers greater flexibility compared to OpenAI’s recent closed systems like GPT-3 and GPT-4.

“We’re excited to see what developers build and how large companies and governments use it where they prefer to run a model themselves,” expressed Altman, sharing optimism about its potential applications.