Qwen

How it grew

Every stage

1. v1 Stage 1

   Qwen

   Qwen-72B

   SIZE72B

   Text

   TextDownloads160K/mo

   TextLikes370

   Language model from Qwen, 72B parameters.

   Open weightsTongyi Qianwen License Agreement257/1040
   +1 more in v1

   • Qwen Turbo

2. v1.5 Stage 2

   Qwen

   Qwen1.5-MoE-A2.7B

   SIZE14B

   Text

   TextDownloads110K/mo

   TextLikes230

   Language model from Qwen, 14B parameters.

   Open weightsTongyi Qianwen license289/1040
   +1 more in v1.5

   • Qwen1.5-1.8B-Chat

3. v2 Stage 3

   Qwen

   Qwen2-7B

   SIZE7.6B

   Text

   TextDownloads210K/mo

   TextLikes180

   Language model from Qwen, 7.6B parameters.

   Open weightsApache 2.0329/1040
   +5 more in v2

   • Qwen2-7B-Instruct
   • Qwen2-1.5B-Instruct
   • Qwen2-1.5B
   • Qwen2-0.5B
   • Qwen2-0.5B-Instruct

4. v2.5 Stage 4

   Qwen

   Qwen2.5 72B Instruct

   CTX131K

   40Quality

   Tools

   TextInput$1.40/M

   TextOutput$5.60/M

   Qwen instruction model for multilingual chat, reasoning, and tool use

   Open weightsQwen license383/1040
   +11 more in v2.5

   • Qwen2.5 32B Instruct
   • Qwen2.5 14B Instruct
   • Qwen2.5 7B Instruct
   • Qwen2.5-7B
   • Qwen2.5-7B-Instruct-1M
   • Qwen2.5-3B-Instruct
   • Qwen2.5-3B
   • Qwen2.5-1.5B-Instruct
   • Qwen2.5-1.5B
   • Qwen2.5-0.5B-Instruct
   • Qwen2.5-0.5B

5. v3 Stage 5

   Qwen

   Qwen3 235B-A22B

   CTX131K

   55Quality

   Reasoning · Tools

   TextInput$0.70/M

   TextOutput$2.80/M

   Large open Qwen MoE for multilingual reasoning, coding, and tool use

   Open weightsApache 2.0524/1040
   +18 more in v3

   • Qwen3 32B
   • Qwen3-Next 80B-A3B Instruct
   • Qwen3 14B
   • Qwen3 8B
   • Qwen3-235B-A22B-Instruct-2507
   • Qwen3-30B-A3B
   • Qwen3-30B-A3B-Instruct-2507
   • Qwen3-14B-Base
   • Qwen3-8B-Base
   • Qwen3-4B
   • Qwen3-4B-Instruct-2507
   • Qwen3-4B-Thinking-2507
   • Qwen3-4B-Base
   • Qwen3-1.7B
   • Qwen3-1.7B-Base
   • Qwen3-0.6B
   • Qwen3-0.6B-Base
   • Qwen3-Next 80B-A3B (Thinking)

6. v3.5 Stage 6

   Qwen

   Qwen3.5 397B-A17B

   CTX262K

   73Quality

   Reasoning · Vision · Tools

   TextImageAudioVideoInput$0.60/M

   TextOutput$3.60/M

   Large open Qwen multimodal MoE for visual agents and long technical tasks

   Open weightsApache 2.0784/1040
   +12 more in v3.5

   • Qwen3.5 122B-A10B
   • Qwen3.5 27B
   • Qwen3.5 35B-A3B
   • Qwen3.5-35B-A3B-Base
   • Qwen3.5-9B
   • Qwen3.5-9B-Base
   • Qwen3.5-4B
   • Qwen3.5-4B-Base
   • Qwen3.5-2B
   • Qwen3.5-2B-Base
   • Qwen3.5-0.8B
   • Qwen3.5-0.8B-Base

7. v3.6 Stage 7

   Qwen

   Qwen3.6 27B

   CTX262K

   73Quality

   Reasoning · Vision · Tools

   TextImageAudioVideoInput$0.60/M

   TextOutput$3.60/M

   Qwen vision-language model for visual reasoning, documents, and agent tasks

   Open weightsApache 2.0887/1040
   +1 more in v3.6

   • Qwen3.6 35B-A3B

8. v3.8 Stage 8

   Qwen

   Qwen3.8-2.4T-A95B

   SIZE2.4T

   Text

   TextDownloads26K/mo

   TextLikes1.3K

   Language model from Qwen, 2.4T parameters.

   Open weightsQwen3.8 Max license971/1040
   +1 more in v3.8

   • Qwen3.8-27B

More Qwen lines

• Qwen Plus 4 stages
• Qwen VL Plus 2 stages
• Qwen Max 5 stages
• Qwen VL 3 stages
• Qwen Coder 2 stages
• Qwen Omni 2 stages
• Qwen Flash 5 stages
• GTE 2 stages
• Qwen Omni 2 stages
• Wan 2 stages
• Qwen Image 2 stages