Category · 12 models
Run the model on the phone, the laptop, or the factory PLC — no round-trip.
Small, quantized models built for latency, privacy and offline use: voice on-device, mobile assistants, robotics control loops, industrial edge.
Microsoft
Small language model punching above its weight on reasoning benchmarks.
Open-weights model family in 1B–27B sizes for on-device & server.
Mistral AI
Fast open-weights small model with strong reasoning.
MiniMax
Open reasoning model with 1M-token context.
Useful Sensors
Open ASR model optimised for real-time edge inference.
Tsinghua University
Visual data comes in various forms, ranging from small icons of just a few pixels to long videos spanning hours.
MiniMax
Today, we are officially open-sourcing and launching MiniMax M2, a model born for Agents and code.
NVIDIA
We present Nemotron 3 Nano 30B-A3B, a Mixture-of-Experts hybrid MambaTransformer language model.
MiniMax
MiniMax's language model tracked by Epoch, focused on chat.
Liquid AI
Liquid AI's second-generation efficient foundation models built on liquid neural networks for on-device use.