Qwen3 VL 30B A3B Thinking vs Qwen3 Coder 480B A35B
Qwen3 VL 30B A3B Thinking is the cheaper option on input by 3%; Qwen3 VL 30B A3B Thinking is cheaper on output. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Qwen3 VL 30B A3B Thinking | Qwen3 Coder 480B A35B | |
|---|---|---|
| Input per 1M tokens | $0.29 | $0.30 |
| Output per 1M tokens | $1 | $1 |
| Cached input per 1M | — | $0.10 |
| Context window | 262K tokens | 262K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | Alibaba | Alibaba |
| Released | — | — |
What the numbers say
- Input price gap: 3% ($0.29 vs $0.30).
- Blended three-to-one input/output rate: $1.87 for Qwen3 VL 30B A3B Thinking, $1.90 for Qwen3 Coder 480B A35B.
- Context: 262K tokens against 262K tokens.
- Licence: Qwen3 VL 30B A3B Thinking is open weights; Qwen3 Coder 480B A35B is open weights.
Sources
Qwen3 VL 30B A3B Thinking specs · Qwen3 Coder 480B A35B specs · Full price index