Twitxy Saved

Ernie Chang

Verified account This account is verified. Learn more

@erniecyc

Research Scientist @Meta AI / Data Curation Lead @ MobileLLM / prev Multimodal Llama
Bay Area deeploid.github.io Joined August 2011
1.5K Following 786 Followers 492 Posts

Latest posts 5

Pinned post

Ernie Chang Verified account @erniecyc
Happy to land this data-efficient model!
Our team is dedicated to building cutting-edge, efficient reasoning models. We are excited to release MobileLLM-R1, a series of sub-billion parameter reasoning models.

Collaborating w/ @zechunliu, Changsheng Zhao et al.
AK Verified account @_akhaliq Sep 12, 2025
Meta just dropped MobileLLM-R1 on Hugging Face

a edge reasoning model with fewer than 1B parameters

2×–5× Performance Boost over other fully open-source models: MobileLLM-R1 achieves ~5× higher MATH accuracy vs. Olmo-1.24B, and ~2× vs. SmolLM2-1.7B.

Uses just 1/10 the pre-training tokens compared to Qwen: matches or surpasses Qwen3 accuracy on multiple reasoning benchmarks while training on only 4.2T tokens (just 11.7% of Qwen3’s 36T).
View quoted post on X
1 replies 5 reposts 34 likes 16K views View on X
Ernie Chang Verified account @erniecyc
Our new 1 billion parameter foundational language model designed to deliver high-quality, efficient on-device inference across a wide range of general language modeling tasks.

Thanks @_akhaliq for sharing.
AK Verified account @_akhaliq Oct 16, 2025
Meta just dropped MobileLLM-Pro on Hugging Face

a 1B foundational language model in the MobileLLM series, designed to deliver high-quality, efficient on-device inference across a wide range of general language modeling tasks

two variants of the model: A pre-trained base model along with quantized checkpoints for CPU and accelerator inference, as well as an instruction tuned version, showing competitive performance against models in the this size range on tasks like tool calling, question answering, rewriting and summarization

MobileLLM-Pro base achieves impressive pre-training results, outperforming Gemma 3 1B and Llama 3.2 1B by on average 5.7% and 7.9% respectively on reasoning, knowledge, and long-context retrieval benchmarks. This performance is achieved by pre-training on less than 2T fully open-source tokens
View quoted post on X
Ernie Chang Verified account @erniecyc
Introducing our latest 1 billion parameter foundational language model designed to deliver high-quality, efficient on-device inference across a wide range of general language modeling tasks.
Patrick Huber Verified account @huberpa91 Oct 16, 2025
1/n 🚀 Introducing MobileLLM-Pro: Our latest 1 billion parameter foundational language model designed to deliver high-quality, efficient on-device inference across a wide range of general language modeling tasks – A thread 🧵

🤗Model Checkpoints: huggingface.co/facebook/Mobil…

Work done with my amazing collaborators Ernie Chang, Wei Wen, Igor Fedorov, Chinnadhurai Sankar, Tarek Elgamal, Hanxian Huang, Naveen Suda, Vish Vogeti, Yanghan Wang, Alex Gladkov, Kai Sheng Tai, Zechun Liu, Abdelrahman Elogeel, Tarek Hefny, Yangyang Shi, Vikas Chandra, Ahmed Aly, Anuj Kumar, Raghuraman Krishnamoorthi, and Adithya Sagar
View quoted post on X
Ernie Chang Verified account @erniecyc
Sharing our latest work for sub-billion LM for on-device scenarios.
AK Verified account @_akhaliq Feb 26, 2024
Meta presents MobileLLM

Optimizing Sub-billion Parameter Language Models for On-Device Use Cases

paper addresses the growing need for efficient large language models (LLMs) on mobile devices, driven by increasing cloud costs and latency concerns. We focus on designing top-quality LLMs with fewer than a billion parameters, a practical choice for mobile deployment. Contrary to prevailing belief emphasizing the pivotal role of data and parameter quantity in determining model quality, our investigation underscores the significance of model architecture for sub-billion scale LLMs. Leveraging deep and thin architectures, coupled with embedding sharing and grouped-query attention mechanisms, we establish a strong baseline network denoted as MobileLLM, which attains a remarkable 2.7%/4.3% accuracy boost over preceding 125M/350M state-of-the-art models. Additionally, we propose an immediate block-wise weight sharing approach with no increase in model size and only marginal latency overhead. The resultant models, denoted as MobileLLM-LS, demonstrate a further accuracy enhancement of 0.7%/0.8% than MobileLLM 125M/350M. Moreover, MobileLLM model family shows significant improvements compared to previous sub-billion models on chat benchmarks, and demonstrates close correctness to LLaMA-v2 7B in API calling tasks, highlighting the capability of small models for common on-device use cases.
View quoted post on X