Qwen3 14B
Alibaba · Released Apr 2025
Qwen3-14B is a 14.8B parameter language model from Alibaba's Qwen3 series with a 40K token context window, designed to handle complex reasoning and dialogue tasks with a configurable thinking mode.
- Strengths
- The model supports switching between a thinking mode for extended reasoning and a direct response mode, allowing it to tackle problem-solving tasks that benefit from deliberation as well as fast dialogue interactions.
- Best for
- Applications requiring a balance between reasoning capability and response speed, such as technical support, multi-step problem solving, or interactive dialogue where you can choose when extended thinking is needed.
- Limitations
- As a smaller dense model, it may struggle with highly specialized domains or very long-context tasks near the 40K token limit compared to larger models, and the extended thinking mode trades latency for reasoning quality.
Input / 1M
$0.2275
Output / 1M
$0.91
Cached input / 1M
—
Context window
131K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in increased 127.5% out increased 279.2%
- 90d
- in increased 127.5% out increased 279.2%
- 1y
- in increased 127.5% out increased 279.2%
- Since launch
- in increased 127.5% out increased 279.2%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 23 Jul 2026 | $0.2275 | $0.91 | — | Imported from OpenRouter | openrouter.ai |
| 19 Jul 2026 | $0.12 | $0.24 | — | Imported from OpenRouter | openrouter.ai |
| 18 Jul 2026 | $0.2275 | $0.91 | — | Imported from OpenRouter | openrouter.ai |
| 17 Jul 2026 | $0.12 | $0.24 | — | Imported from OpenRouter | openrouter.ai |
| 16 Jul 2026 | $0.12 | $0.24 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.1 | $0.24 | — | Imported from OpenRouter | openrouter.ai |