Qwen
Qwen 3 0.6B
Smallest dense Qwen3 release with switchable thinking and non-thinking modes in a very light deployment footprint.
Overview and architecture
What it is
Company
Family
Release date
Architecture
License
Modality
Context window
Total params
Active params
Layers
Hidden size
Attention heads
KV heads
KV-bearing layers
Research highlight
What improved
Thinking-mode switch
Qwen3’s defining change is seamless switching between deeper reasoning and faster non-thinking dialogue within the same checkpoint.
Reasoning uplift
Qwen positions the line as stronger than QwQ in thinking mode and stronger than Qwen2.5 instruct models in non-thinking mode on reasoning-heavy tasks.
Agent and multilingual focus
The release also emphasizes stronger agent use and support for 100+ languages and dialects, even at smaller sizes.
Training and release context
How it was released
Family release
Qwen3 is released as a dense and MoE model family centered on switching between thinking and non-thinking modes within the same model.
Training stage
Qwen describes the release as a pretraining plus post-training model rather than a small instruction-only adaptation.
Context packaging
The 0.6B model is published with 32K native context, and the larger dense variants explicitly extend to 131K with YaRN.
Where it is strong
Where it is strong
Thinking and non-thinking use
The 0.6B release is built to switch between deeper reasoning mode and faster general dialogue mode without changing models.
Agent workflows
Qwen positions the family for tool use and agent-style tasks in both thinking and non-thinking modes.
Multilingual assistant work
The family is published with support for 100+ languages and dialects, making it a broad multilingual assistant line rather than a narrow specialist release.
Memory behavior
What dominates VRAM
At this size the resident weight floor stays small, so runtime reserve and long-context cache can become a larger fraction of total VRAM than on bigger dense models.
Sources