Qwen3 235B A22B Thinking 2507
Alibaba · Released Jul 2025
A 235-billion-parameter mixture-of-experts model from Alibaba that activates 22 billion parameters per token, with extended reasoning capabilities and a 131k-token context window.
- Strengths
- Delivers reasoning depth across a large parameter count while keeping per-token computation efficient through sparse activation.
- Best for
- Complex problem-solving and multi-step reasoning tasks where model scale and inference speed both matter.
- Limitations
- Reasoning overhead makes it slower than the non-thinking variant Qwen3 235B A22B Instruct 2507; the 80-billion-parameter Qwen3 Next 80B A3B Thinking offers extended reasoning with a larger context window.
Input / 1M
$0.45
Output / 1M
$3.50
Cached input / 1M
—
Context window
128K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in increased 95.7% out increased 52.2%
- 90d
- in increased 50.0% out increased 16.7%
- 1y
- in increased 350.0% out increased 3400.0%
- Since launch
- in increased 350.0% out increased 3400.0%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 9 Oct 2026 | $0.45 | $3.50 | — | Imported from OpenRouter | openrouter.ai |
| 31 Jul 2026 | $0.23 | $2.30 | — | Imported from OpenRouter | openrouter.ai |
| 20 Jul 2026 | $0.3 | $3.00 | — | Imported from OpenRouter | openrouter.ai |
| 1 Jul 2026 | $0.1495 | $1.50 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.1 | $0.1 | $0.1 | Imported from OpenRouter | openrouter.ai |