| AI model | Parameters | Active parameters | Layers | KV heads | Context |
| Qwen (Alibaba Cloud) Qwen-7B | 7.72 B | – | 32 layers | – | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen-7B-Chat | 7.72 B | – | 32 layers | – | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-0.5B | 0.62 B | – | 24 layers | 16 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-0.5B-Chat | 0.62 B | – | 24 layers | 16 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-1.8B-Chat | 1.84 B | – | 24 layers | 16 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-7B | 7.72 B | – | 32 layers | 32 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-32B-Chat | 32.5 B | – | 64 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-MoE-A2.7B | 14.3 B | – | 24 layers | 16 heads | 8,192 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen1.5-MoE-A2.7B-Chat | 14.3 B | – | 24 layers | 16 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-0.5B | 0.49 B | – | 24 layers | 2 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-0.5B-Instruct | 0.49 B | – | 24 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-1.5B | 1.54 B | – | 28 layers | 2 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-1.5B-Instruct | 1.54 B | – | 28 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-7B | 7.62 B | – | 28 layers | 4 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-7B-Instruct | 7.62 B | – | 28 layers | 4 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2-72B-Instruct | 72.7 B | – | 80 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5 7B Instruct | 7.61 B | – | 28 layers | 4 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5 14B Instruct | 14.7 B | – | 48 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5 32B Instruct | 32.5 B | – | 64 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5 72B Instruct | 72.7 B | – | 80 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-0.5B | 0.49 B | – | 24 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-0.5B-Instruct | 0.49 B | – | 24 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-1.5B | 1.54 B | – | 28 layers | 2 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-1.5B-Instruct | 1.54 B | – | 28 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-3B | 3.09 B | – | 36 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-3B-Instruct | 3.09 B | – | 36 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-7B | 7.62 B | – | 28 layers | 4 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-7B-Instruct-1M | 7.62 B | – | 28 layers | 4 heads | 1,010,000 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-14B | 14.8 B | – | 48 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-14B-Instruct-1M | 14.8 B | – | 48 layers | 8 heads | 1,010,000 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-0.5B | 0.49 B | – | 24 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-0.5B-Instruct | 0.49 B | – | 24 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-1.5B | 1.54 B | – | 28 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-1.5B-Instruct | 1.54 B | – | 28 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-3B | 3.09 B | – | 36 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-3B-Instruct | 3.09 B | – | 36 layers | 2 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-7B | 7.62 B | – | 28 layers | 4 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-7B-Instruct | 7.62 B | – | 28 layers | 4 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-14B-Instruct | 14.8 B | – | 48 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Coder-32B-Instruct | 32.8 B | – | 64 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Math-1.5B | 1.54 B | – | 28 layers | 2 heads | 4,096 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Math-1.5B-Instruct | 1.54 B | – | 28 layers | 2 heads | 4,096 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Math-7B | 7.62 B | – | 28 layers | 4 heads | 4,096 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen2.5-Math-7B-Instruct | 7.62 B | – | 28 layers | 4 heads | 4,096 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3 8B | 8.2 B | – | 36 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3 14B | 14.8 B | – | 40 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3 30B-A3B | 30.5 B | 3.3 B | 48 layers | 4 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3 32B | 32.8 B | – | 64 layers | 8 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3 235B-A22B | 235 B | 22 B | 94 layers | 4 heads | 131,072 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-0.6B | 0.75 B | – | 28 layers | 8 heads | 40,960 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-0.6B-Base | 0.6 B | – | 28 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-1.7B | 2.03 B | – | 28 layers | 8 heads | 40,960 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-1.7B-Base | 1.72 B | – | 28 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-4B | 4.02 B | – | 36 layers | 8 heads | 40,960 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-4B-Base | 4.02 B | – | 36 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-4B-Instruct-2507 | 4.02 B | – | 36 layers | 8 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-4B-Thinking-2507 | 4.02 B | – | 36 layers | 8 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-8B-Base | 8.19 B | – | 36 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-14B-Base | 14.8 B | – | 40 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-30B-A3B-Base | 30.5 B | – | 48 layers | 4 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-30B-A3B-Instruct-2507 | 30.5 B | – | 48 layers | 4 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-30B-A3B-Thinking-2507 | 30.5 B | – | 48 layers | 4 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-235B-A22B-Instruct-2507 | 235 B | – | 94 layers | 4 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-Coder-30B-A3B-Instruct | 30.5 B | – | 48 layers | 4 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-Coder-480B-A35B-Instruct | 480 B | – | 62 layers | 8 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-Coder-Next | 79.7 B | – | 48 layers | 2 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-Next-80B-A3B-Instruct | 81.3 B | – | 48 layers | 2 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3-Next-80B-A3B-Thinking | 81.3 B | – | 48 layers | 2 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3.8-2.4T-A95B | 2,446 B | – | 92 layers | 4 heads | 262,144 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3Guard-Gen-0.6B | 0.75 B | – | 28 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3Guard-Gen-4B | 4.41 B | – | 36 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) Qwen3Guard-Gen-8B | 8.19 B | – | 36 layers | 8 heads | 32,768 tokens |
|---|
| Qwen (Alibaba Cloud) QwQ-32B | 32.8 B | – | 64 layers | 8 heads | 40,960 tokens |