# deepseek-v3.2-thinking

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-medal results on the 2025 IMO and IOI. V3.2 also uses a large-scale agentic task synthesis pipeline to better integrate reasoning into tool-use settings, boosting compliance and generalization in interactive environments.

Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config)

## Model Information

- **Organization**: [DeepSeek](/llm.txt)
- **Slug**: deepseek-v3-2-thinking
- **Available at Providers**: 25
- **Release Date**: December 1, 2025

### Benchmark Scores
- Weekly: 6.93

## Providers

| Provider | Name | $ Input (per 1M) | $ Output (per 1M) | Free | Link |
|----------|------|-----------------|------------------|------|------|
| [Vercel AI Gateway](/llm/vercel.txt) | DeepSeek V3.2 Thinking | 0.28 | 0.42 |  |  |
| [302.AI](/llm/302ai.txt) | deepseek-v3.2-thinking | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-207705121) |
| [302.AI](/llm/302ai.txt) | deepseek/deepseek-v3.2 | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-308032503) |
| [302.AI](/llm/302ai.txt) | sophnet/DeepSeek-V3.2 | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-319727732) |
| [302.AI](/llm/302ai.txt) | Pro/deepseek-ai/DeepSeek-V3.2 | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-252564719) |
| [302.AI](/llm/302ai.txt) | deepseek-ai/DeepSeek-V3.2 | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-252564719) |
| [Arena AI](/llm/arenaai.txt) |  |  |  |  |  |
| [Arena AI](/llm/arenaai.txt) |  |  |  |  |  |
| [Yupp](/llm/yupp.txt) | DeepSeek V3.2 (Amazon Bedrock) |  |  |  |  |
| [Together AI](/llm/togetherai.txt) | Deepseek V3.2 | 0.00 | 0.00 |  |  |
| [Okara](/llm/okara.txt) | Deepseek V3.2 |  |  |  | [View](https://okara.ai/ai-models/deepseek-v3.2) |
| [StreamLake](/llm/streamlake.txt) | DeepSeek-V3.2 | 0.29 | 0.43 |  |  |
| [CommonStack](/llm/commonstack.txt) | DeepSeek: DeepSeek V3.2 | 0.27 | 0.40 |  |  |
| [NetMind](/llm/netmind.txt) | DeepSeek-V3.2 | 0.27 | 0.40 |  |  |
| [Blackbox AI](/llm/blackboxai.txt) | blackboxai/deepseek/deepseek-v3.2 |  |  |  |  |
| [CometAPI](/llm/cometapi.txt) | DeepSeek-V3.2 | 0.22 | 0.35 |  |  |
| [MegaNova](/llm/meganova.txt) | DeepSeek V3.2 | 0.26 | 0.38 |  |  |
| [Qiniu](/llm/qiniuai.txt) | Deepseek/DeepSeek-V3.2 |  |  |  |  |
| [LangDB](/llm/langdb.txt) | deepseek-v3.2 |  |  |  | [View](https://langdb.ai/app/models) |
| [ApiYI](/llm/apiyi.txt) | deepseek-v3.2 |  |  |  |  |
| [WaveSpeed AI](/llm/wavespeed.txt) | deepseek-v3.2 | 0.28 | 0.42 |  |  |
| [Airforce API](/llm/airforce.txt) | deepseek-v3.2 |  |  |  |  |
| [Airforce API](/llm/airforce.txt) | deepseek-v3.2-thinking |  |  |  |  |
| [Writingmate](/llm/writingmate.txt) | DeepSeek: DeepSeek V3.2 |  |  |  | [View](https://writingmate.ai/models/deepseek/deepseek-v3.2) |
| [SiliconFlow (China)](/llm/siliconflowcn.txt) | Pro/deepseek-ai/DeepSeek-V3.2 | 0.27 | 0.42 |  |  |

---

[← Back to all providers](/llm.txt)