New open model from Tencent Hy: Hy3 (295B total 21B active - apache 2.0)
Tencent has released its new open model, Hy3, with 295B total parameters and 21B active parameters. The model is licensed under Apache 2.0.
- Tencent has released its new open model, Hy3, with 295B total parameters and 21B active parameters
- The model is licensed under Apache 2.0, making it more accessible for global use
- The release of Hy3 provides an alternative to other large language models
- The model is available on the Hugging Face platform
The Hy3 model is a significant release from Tencent, offering a large-scale language model with 295B total parameters and 21B active parameters.
The model is now licensed under Apache 2.0, which is a more permissive license compared to the previous community license. This change in licensing is expected to make the model more accessible for use in various regions, including South Korea, the UK, and the EU.
The release of Hy3 is notable, as it provides an alternative to other large language models, and its open nature allows for community involvement and development. The model is available on the Hugging Face platform, making it easily accessible for developers and researchers.
The community's reaction to the release is positive, with many users appreciating the improved licensing and the potential for collaboration and innovation that the model offers.
Source: New open model from Tencent Hy: Hy3 (295B total 21B active - apache 2.0). Read the full piece at the source.
offers a new open model for development and research
provides an alternative to other large language models
offers a valuable resource for learning and research
advances the field of natural language processing
New model: GigaChat3.5-432B-A28B (with day-0 GGUF support!)
Doing the actual math on a $20k local AI rig breakeven
