CoolFace
Apppublic

TESTforge42/smollm3-3b-cpu

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face
0likes
App README

Nomos42 SmolLM3-3B CPU

OpenAI-compatible inference for SmolLM3-3B (GGUF Q4KM, ~1.9 GB). 128K context, dual think/no-think mode, 6 languages. Released 2026.

MMLU 68.9 · beats Llama-3.2-3B (66.5) · close to Qwen3-4B (69.4)

  • —GET / → {model, ready, ...}
  • —POST /v1/chat/completions

<!-- touch: 2026-04-17T20:50:23.682702Z -->