Qwen3 0.6BF16
requires 5 GB+ memory · 33K context · Phone / edge
1.5 GBWeights
← All models4K 1.4 GB 8K 1.5 GB 16K 1.5 GB 33K 1.6 GB
Qwen3
Think when you need it, answer fast when you don’t. Qwen3 switches thinking/non-thinking modes for coding, math, and 119 languages, from 0.6B on a phone to 30B-A3B MoE on a workstation.
- Maker
- Alibaba
- Architecture
- Hybrid reasoning · dense & MoE
- Fits from
- Phone / edge · 4 GB+ RAM
- Context
- up to 33K
- Runs
- Local Intelligence
ReasoningCodingMultilingualMath & code119 languagesThink / fast modes
Sizes
Qwen3 1.7BF16
requires 4 GB+ memory · 33K context · Phone / edge
1.1 GBWeights
Qwen3 4BQ4_K_M
requires 6 GB+ memory · 33K context · Phone / edge
2.5 GBWeights
Qwen3 8BQ4_K_M
requires 10 GB+ memory · 33K context · Laptop
5 GBWeights
Qwen3 14BQ4_K_M
requires 16 GB+ memory · 33K context · Desktop
9 GBWeights
Qwen3 30B-A3BQ4_K_M
requires 32 GB+ memory · 33K context · Workstation
20 GBWeights
Memory at context length
Approximate working set for Qwen3 1.7B as context grows.