Skip to content

qwen3.5

Qwen3.5, a stock instruct model read zero-shot in the chat layout.

ardana pull qwen3.5

qwen3.5:0.8b latest

Parameters: 0.8B
Released: Feb 28, 2026
Base model: Qwen/Qwen3.5-0.8B
Weights: Hugging Face

Tags
Tag	Size	Pull command
qwen3.5:0.8b default	563 MB	`ardana pull qwen3.5:0.8b`
qwen3.5:0.8b-q8_0	834 MB	`ardana pull qwen3.5:0.8b-q8_0`
qwen3.5:0.8b-bf16	1.6 GB	`ardana pull qwen3.5:0.8b-bf16`

qwen3.5:2b

Parameters: 2B
Released: Feb 28, 2026
Base model: Qwen/Qwen3.5-2B
Weights: Hugging Face

Tags
Tag	Size	Pull command
qwen3.5:2b default	1.3 GB	`ardana pull qwen3.5:2b`
qwen3.5:2b-q8_0	2.0 GB	`ardana pull qwen3.5:2b-q8_0`
qwen3.5:2b-bf16	3.8 GB	`ardana pull qwen3.5:2b-bf16`

qwen3.5:4b

Parameters: 4B
Released: Feb 27, 2026
Base model: Qwen/Qwen3.5-4B
Weights: Hugging Face

Tags
Tag	Size	Pull command
qwen3.5:4b default	2.7 GB	`ardana pull qwen3.5:4b`
qwen3.5:4b-q8_0	4.5 GB	`ardana pull qwen3.5:4b-q8_0`
qwen3.5:4b-bf16	8.4 GB	`ardana pull qwen3.5:4b-bf16`

qwen3.5:9b

Parameters: 9B
Released: Feb 27, 2026
Base model: Qwen/Qwen3.5-9B
Weights: Hugging Face

Tags
Tag	Size	Pull command
qwen3.5:9b default	5.7 GB	`ardana pull qwen3.5:9b`
qwen3.5:9b-q8_0	9.5 GB	`ardana pull qwen3.5:9b-q8_0`
qwen3.5:9b-bf16	17.9 GB	`ardana pull qwen3.5:9b-bf16`