PrismML has released Bonsai 27B, a 27B-class AI model capable of running on smartphones, including iPhones. Utilizing 1-bit weights, the model fits within the device's memory constraints, marking a significant leap in mobile AI technology. This development highlights the growing potential for advanced AI applications on consumer devices, possibly expanding AI accessibility.
Bonsai 27B, developed by PrismML, represents a groundbreaking step in AI models as it becomes the first 27B-class AI model capable of running on smartphones, including iPhones. This model leverages low-bit weight representations, allowing it to fit within the memory constraints of mobile devices.
The introduction of Bonsai 27B involves two variants: the 1-bit and the ternary models. The 1-bit variant with binary weights packs the entire model in approximately 4 GB. This efficient compression allows it to operate within the limited memory environments of smartphones without compromising on performance.
Bonsai 27B is capable of running complex tasks such as multi-step reasoning, structured tool calls, and vision tasks traditionally conducted on high-capacity machines. By making such capabilities available on smartphones, the model could democratize AI technology, providing broader access to advanced AI functionalities.
The release also ties into PrismML's reported discussions with Apple regarding potential collaborations, underlining the possible integration of advanced AI technology in everyday consumer electronics. This move aligns with growing trends in making AI more accessible and integrated into personal devices.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
PrismML has released its Bonsai 27B artificial intelligence model, claiming it is the first of its size capable of running on iPhone, iPad, and Mac. The model is notable for operating natively on Apple hardware and includes a developer preview API, coinciding with reported discussions between PrismML and Apple regarding the use of its technology.
Bonsai 27B, based on Qwen3.6, is the first 27B-class model capable of running on smartphones. It utilizes novel low-bit weight representations, enabling complex tasks such as multi-step reasoning and vision processing on devices with limited memory. This advancement could significantly impact AI accessibility and usability in mobile technology.