Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...
| Field | Per token (USD) | Per million |
|---|---|---|
| prompt | 0.00000008 | $0.080 |
| completion | 0.00000028 | $0.280 |