SmolLM2 360M Instruct
PermissiveHugging Face · SmolLM2 · Released 2024-11
360MText
A sub-billion-parameter model small enough to run in a browser via WebGPU, useful for constrained on-device tasks rather than general chat.
Strengths
- +Apache 2.0 license
- +Runs in-browser and on microcontroller-class hardware
- +Extremely low latency
Limitations
- -Limited to simple tasks; not a general-purpose chat replacement
- -Frequent factual errors on anything beyond basic knowledge
License
Apache 2.0
use commercially with attribution niceties
Hardware
runs on a good consumer GPU (or Apple Silicon) with quantization
Links
Stats
HuggingFaceTB/SmolLM2-360M-Instruct— downloads·— likes
via Hugging Face