Research
65 updates on Research.
Meta's MobileLLM-Pro runs a 128k context from a 590 MB 4-bit build
Meta Reality Labs released a 1.08B on-device model with a 128k context window, measured at 33.6 tok/s decode on a Galaxy S25 CPU.
Confidant fine-tunes Phi2-2.7B across a phone and two laptops in 40.1 hours
Zhejiang University researchers split a language model across three everyday devices and fine-tuned Phi2-2.7B on Alpaca in 40.1 hours without moving raw data.
lm-Meter times on-device inference and finds prefill, not decode, is the bottleneck
A profiler that runs inside the phone measures each inference phase and reports prefill slowing 158 times as models grow, against 10 times for decode.
Meta trains 140M to 950M reasoning models on 4.2T tokens
MobileLLM-R1 spans 140M to 950M parameters, trained on 4.2T tokens, and Meta scores the 950M model at 74.0 on MATH500.
ShadowNPU scores attention on the Snapdragon NPU and reports 4.5 times faster inference
Peking University and BUPT move attention token scoring onto a Snapdragon NPU and report up to 4.5 times faster inference using a single CPU core.
P/D-Device prefills in the cloud and decodes on the phone, cutting TTFT 60%
Huawei researchers split prefill and decoding between cloud and phone, reporting time to first token down at least 60% and cloud throughput up to 15x.
Benchmark of 68 small language models finds architecture outweighs size on device
A study of 68 models from 100M to 5B puts Phi-3 near 70 percent accuracy and finds first-token time and memory tracking architecture, not parameter count.
Vijay Janapa Reddi argues edge generative AI needs models under 1B parameters
In ACM Queue, Vijay Janapa Reddi argues generative AI reaches edge devices only through small deployable models, with sub-1B ones the viable case.
Alibaba gives Qwen3's 0.6B and 1.7B models a reasoning switch
Qwen3-0.6B and Qwen3-1.7B carry the family's switch between a reasoning mode and a fast mode, with 32K context and Apache 2.0 weights.
Google's ML Drift runs Llama 3.1 8B on a phone GPU at 12.7 tokens per second
The paper measures 37.1 decode tokens per second for Gemma2 2B and 12.7 for Llama 3.1 8B on the Adreno 750 GPU of a Samsung S24.
D2MoE picks a bit-width per token, 1.39 times the throughput at up to 53 percent less memory
A MobiCom 2025 paper routes every token to an expert and a bit-width, reporting up to 1.39 times the throughput of EdgeMoE at up to 53 percent less memory.
ROMA keeps a 4-bit 3B model in on-chip ROM and reports 31,800 tok/s in synthesis
A synthesised 7 nm accelerator holds a 4-bit 3B Llama in read-only memory and the LoRA adapter in SRAM, with no chip and no FPGA prototype built.