Qwen 3 0.6B
by Alibaba · qwen-3 family
0.6B
parameters
text-generation reasoning multilingual
Qwen 3 0.6B is the smallest model in Alibaba's Qwen 3 family, designed for ultra-lightweight deployment on resource-constrained devices. Despite its compact size, it supports reasoning and multilingual capabilities with a 32K context window. This model is ideal for edge deployment, mobile devices, and scenarios where minimal VRAM and fast inference are priorities. It can run comfortably on virtually any modern GPU or Apple Silicon Mac.
Quick Start with Ollama
ollama run 0.6b-q4_K_M | Creator | Alibaba |
| Parameters | 600M |
| Architecture | transformer-decoder |
| Context | 32K tokens |
| Released | Apr 29, 2025 |
| License | Apache 2.0 |
| Ollama | qwen3:0.6b |
Quantization Options
| Format | File Size | VRAM Required | Quality | Ollama Tag |
|---|---|---|---|---|
| Q4_K_M rec | 0.5 GB | 2.5 GB | | 0.6b-q4_K_M |
| Q8_0 | 0.7 GB | 2.7 GB | | 0.6b-q8_0 |
| F16 | 1.3 GB | 3.3 GB | | 0.6b-fp16 |
Compatible Hardware
Q4_K_M requires 2.5 GB VRAM