Qwen3 235B A22B Instruct 2507 vs Qwen3 30B A3B
Qwen3 235B A22B Instruct 2507 is the cheaper option on input by 25%; Qwen3 30B A3B is cheaper on output by 9%. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Qwen3 235B A22B Instruct 2507 | Qwen3 30B A3B | |
|---|---|---|
| Input per 1M tokens | $0.09 | $0.12 |
| Output per 1M tokens | $0.55 | $0.50 |
| Cached input per 1M | — | — |
| Context window | 262K tokens | 41K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | Alibaba | Alibaba |
| Released | — | — |
What the numbers say
- Input price gap: 25% ($0.09 vs $0.12).
- Blended three-to-one input/output rate: $0.82 for Qwen3 235B A22B Instruct 2507, $0.86 for Qwen3 30B A3B.
- Context: 262K tokens against 41K tokens.
- Licence: Qwen3 235B A22B Instruct 2507 is open weights; Qwen3 30B A3B is open weights.
Sources
Qwen3 235B A22B Instruct 2507 specs · Qwen3 30B A3B specs · Full price index