Qwen3 235B A22B 2507 is cheaper than Qwen3 Max Thinking at $0.00015/M vs $1.2/M input tokens.
| Spec | Qwen3 235B A22B 2507 | Qwen3 Max Thinking |
|---|---|---|
| Input price | $0.00015/M tokens | $1.2/M tokens |
| Output price | $0.00085/M tokens | $6/M tokens |
| Context window | — | 262K tokens |
| Released | Jul 2025 | Feb 2026 |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 235B A22B 2507 costs 7059x less per token.
Qwen3 235B A22B 2507 is cheaper on both — 8000× input, 7059× output
Qwen3 235B A22B 2507 uses 3.2x more transitions
Qwen3 235B A22B 2507 is cheaper than Qwen3 Max Thinking at $0.00015/M vs $1.2/M input tokens.
| Spec | Qwen3 235B A22B 2507 | Qwen3 Max Thinking |
|---|---|---|
| Input price | $0.00015/M tokens | $1.2/M tokens |
| Output price | $0.00085/M tokens | $6/M tokens |
| Context window | — | 262K tokens |
| Released | Jul 2025 | Feb 2026 |
Not enough votes to call it. On the specs, nothing separates them.
Qwen3 235B A22B 2507 costs 7059x less per token.
Qwen3 235B A22B 2507 is cheaper on both — 8000× input, 7059× output
Qwen3 235B A22B 2507 uses 3.2x more transitions