Apple releases eight open-source language models named OpenELM for on-device AI capabilities in iOS 18.
Models available on Hugging Face Hub include four pre-trained models using CoreNet library and four instruction-tuned models.
OpenELM initiative expected to bring advancements in AI capabilities, address privacy concerns, and provide faster response times compared to cloud-based models.
In a significant development, Apple has released eight open-source language models named OpenELM that are designed to run on devices rather than through cloud servers. This move by the tech giant is expected to bring artificial intelligence capabilities to its devices in iOS 18. The OpenELM models are available on the Hugging Face Hub and include four pre-trained models using the CoreNet library and four instruction-tuned models. Apple has also provided complete frameworks for training and evaluation of the language model on publicly available datasets, along with training logs, multiple checkpoints, and pre-training configurations. This initiative by Apple is expected to empower the open research community with state-of-the-art language models while addressing concerns around privacy and faster response times compared to cloud-based models. In addition to this, Apple has also submitted research papers on the techniques used for efficient language models and has opened up all of the training data, evaluation metrics, checkpoints, and training configurations for public access. This move by Apple is expected to bring significant advancements in AI capabilities on devices and is a major step towards making AI more accessible to users while ensuring privacy and faster response times.
Apple is planning to improve the next iPhone 16 and iPhone 16 Pro with artificial intelligence.
,Apple is expected to use more efficient processing techniques and optimized hardware for generative AI in the next generation of Axx chips.
,Apple has published a research paper on the techniques used for efficient language models and has open-sourced all of the training data, evaluation metrics, checkpoints, and training configurations.
,Apple is working with Cornell University on Ferret-UI, an LLM that helps understand a device’s user interface and offers numerous interactions such as voice navigation or describing screen content for those with impaired vision.
,Apple has submitted eight large language models to the Hugging Face hub, an online resource for open-source AI implementations.
Apple is developing a new LLM that is expected to run on-device.
An on-device-based LLM means it will depend on the iPhone processor rather than the cloud, resulting in better performance.
Apple may license Google’s Gemini AI software to handle complex tasks requiring an internet connection.
Apple's LLM will address concerns about privacy and faster response times compared to cloud-based models like ChatGPT and Copilot.
Accuracy
No Contradictions at Time
Of
Publication
Deception
(100%)
None Found At Time Of
Publication
Fallacies
(100%)
None Found At Time Of
Publication
Bias
(95%)
The author expresses a bias towards Apple's new LLM by stating that it will outperform OpenAI's GPT-4 and Microsoft's Copilot, without providing any evidence to support this claim. The author also mentions the potential copyright infringement issues with Google's Gemini AI software being used in Apple's LLM, but does not provide any context or evidence regarding these lawsuits.
Apple is developing its LLM that will reportedly run on-device to affirm the consumers’ privacy while simultaneously promising great performance.
It’ll also be interesting to see how the model competes against ChatGPT, which currently holds a huge percentage of the mobile market share...
The model will spot better performance...
While reports indicate Apple’s LLM will outperform OpenAI’s GPT-4 model...
Apple released eight open source large language models (LLMs) named OpenELM.
,OpenELM models are designed to run on-device instead of through cloud servers.
,Apple uses a layer-wise scaling strategy for improving accuracy and efficiency in OpenELM.
,OpenELM exhibits a 2.36% improvement in accuracy compared to OLMo while requiring 2x fewer pre-training tokens.
Apple released the complete framework for training and evaluation of the language model on publicly available datasets along with training logs, multiple checkpoints, and pre-training configurations.
,OpenELM is aimed at empowering and enriching the open research community with state-of-the-art language models.
,Apple plans to bring AI capabilities to its devices in iOS 18
Apple announced a special event on May 7, 2024, for unveiling new products or features.
Accuracy
OpenELM models are designed to run on-device instead of through cloud servers.
Apple uses a layer-wise scaling strategy for improving accuracy and efficiency in OpenELM.
