Model Intelligence Sheet
1bit-MONSTER/SmolLM3-3B-GGUF overview
SmolLM3 3B — GGUF Q4 K M llama.cpp project's own Q4 K M GGUF, re hosted with measured performance for the 1bit engine https://github.com/1bit MONSTER/engine on…
Runs locally from ~1.78 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
1 GGUF files detected
Direct downloads for local inference
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| SmolLM3-Q4_K_M.gguf | GGUF | Q4_K_M | 1.78 GB | Download |
Model Details
| Model ID | 1bit-MONSTER/SmolLM3-3B-GGUF |
|---|---|
| Author | 1bit-MONSTER |
| Pipeline | — |
| License | apache-2.0 |
| Base model | HuggingFaceTB/SmolLM3-3B |
| Last modified | 2026-09-26T15:39:31.000Z |
Model README
---
license: apache-2.0
base_model:
- HuggingFaceTB/SmolLM3-3B
tags:
- gguf
---
SmolLM3-3B — GGUF (Q4_K_M)
llama.cpp project's own Q4_K_M GGUF, re-hosted with measured performance for the
1bit engine on Strix Halo.
Contents
SmolLM3-Q4_K_M.gguf
Measured performance (Strix Halo, Vulkan)
pp512: 3003 tok/s · tg128: 93.1 tok/s
Running it
1bit serve -m SmolLM3-Q4_K_M.gguf --device vulkanAttribution
- Model: HuggingFaceTB/SmolLM3-3B, Apache 2.0.
- Quantization: ggml-org/SmolLM3-3B-GGUF.
- License: Apache 2.0, inherited from the base model.
Run 1bit-MONSTER/SmolLM3-3B-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models