Qwen3 14B vs Qwen3 30B A3B Instruct 2507
Qwen3 30B A3B Instruct 2507 is the cheaper option on input by 17%; Qwen3 14B is cheaper on output by 20%. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Qwen3 14B | Qwen3 30B A3B Instruct 2507 | |
|---|---|---|
| Input per 1M tokens | $0.12 | $0.10 |
| Output per 1M tokens | $0.24 | $0.30 |
| Cached input per 1M | — | $0.10 |
| Context window | 41K tokens | 262K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | Alibaba | Alibaba |
| Released | — | — |
What the numbers say
- Input price gap: 17% ($0.12 vs $0.10).
- Blended three-to-one input/output rate: $0.60 for Qwen3 14B, $0.60 for Qwen3 30B A3B Instruct 2507.
- Context: 41K tokens against 262K tokens.
- Licence: Qwen3 14B is open weights; Qwen3 30B A3B Instruct 2507 is open weights.
Sources
Qwen3 14B specs · Qwen3 30B A3B Instruct 2507 specs · Full price index