Fully open 3B language model with switchable reasoning mode, long context, tool calling and multilingual support, released with its training recipe.
Qwen/Qwen3-0.6B
Indexed from Hugging FaceLinked to its home on Hugging Face. FrontierOS indexes it; nothing is hosted or mirrored here.ModelSmall language model751M paramsEst. Single-board computerEstimated minimum compute for inference: Raspberry Pi 5 class CPU. Not a benchmark.Apache 2.0Desk device
About
Smallest dense model in the Qwen3 family, with optional thinking mode and function-calling support.
Why it matters for physical AI
Small enough for a Raspberry Pi-class host to handle intent parsing and simple tool calls offline.
Summary written by FrontierOS in our own words. The upstream card and license govern use. Metadata checked 27 Sep 2026. Read the full card on Hugging Face
Files, weights and version history live on Hugging Face. FrontierOS indexes this repo; it does not host or mirror it.
hf download Qwen/Qwen3-0.6BUses the Hugging Face CLI (pip install -U huggingface_hub). This downloads the whole repo; add --include "<pattern>" to fetch only the files you need.
What's the lowest-latency fully local voice loop you've built?
A typical cascade is voice activity detection, a turn detector, streaming speech-to-text, a small language model and a fast text-to-speech voice.
More small language model
Browse allTiny instruction-tuned Gemma 3 model meant to be fine-tuned for narrow jobs such as classification, extraction or command routing.