Qwen3 30B A3B Thinking 2507 is cheaper than DeepSeek V3.2 Exp at $0.071/M vs $0.2/M input tokens.
| Spec | DeepSeek V3.2 Exp | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
| Input price | $0.2/M tokens | $0.071/M tokens |
| Output price | $0.4/M tokens | $0.285/M tokens |
| Context window | 164K tokens | 262K tokens |
| Released | Sep 2025 | Aug 2025 |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 30B A3B Thinking 2507 is cheaper on both — 2.8× input, 1.4× output
Qwen3 30B A3B Thinking 2507 uses 44.1x more emoji
Qwen3 30B A3B Thinking 2507 is cheaper than DeepSeek V3.2 Exp at $0.071/M vs $0.2/M input tokens.
| Spec | DeepSeek V3.2 Exp | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
| Input price | $0.2/M tokens | $0.071/M tokens |
| Output price | $0.4/M tokens | $0.285/M tokens |
| Context window | 164K tokens | 262K tokens |
| Released | Sep 2025 | Aug 2025 |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 30B A3B Thinking 2507 is cheaper on both — 2.8× input, 1.4× output
Qwen3 30B A3B Thinking 2507 uses 44.1x more emoji