Qwen3 Coder 30B A3B Instruct vs Qwen2.5 7B Instruct
Qwen3 Coder 30B A3B Instruct is the cheaper option on input by 30%; Qwen2.5 7B Instruct is cheaper on output by 29%. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Qwen3 Coder 30B A3B Instruct | Qwen2.5 7B Instruct | |
|---|---|---|
| Input per 1M tokens | $0.07 | $0.10 |
| Output per 1M tokens | $0.28 | $0.20 |
| Cached input per 1M | — | — |
| Context window | 262K tokens | 33K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | Alibaba | Alibaba |
| Released | — | — |
What the numbers say
- Input price gap: 30% ($0.07 vs $0.10).
- Blended three-to-one input/output rate: $0.49 for Qwen3 Coder 30B A3B Instruct, $0.50 for Qwen2.5 7B Instruct.
- Context: 262K tokens against 33K tokens.
- Licence: Qwen3 Coder 30B A3B Instruct is open weights; Qwen2.5 7B Instruct is open weights.
Sources
Qwen3 Coder 30B A3B Instruct specs · Qwen2.5 7B Instruct specs · Full price index