# deepseek-v3.2

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
- **Available at Providers**: 44

## Providers

| Provider | Name | $ Input (per 1M) | $ Output (per 1M) | Free | Link |
|----------|------|-----------------|------------------|------|------|
| [AIHubMix](/llm/aihubmix.txt) | deepseek-v3.2-fast | 1.10 | 3.29 |  | [View](https://aihubmix.com/model/deepseek-v3.2-fast) |
| [AIHubMix](/llm/aihubmix.txt) | deepseek-v3.2 | 0.30 | 0.45 |  | [View](https://aihubmix.com/model/deepseek-v3.2) |
| [AIHubMix](/llm/aihubmix.txt) | deepseek-v3.2-think | 0.30 | 0.45 |  | [View](https://aihubmix.com/model/deepseek-v3.2-think) |
| [FastRouter](/llm/fastrouter.txt) | DeepSeek: DeepSeek V3.2 | 0.27 | 0.40 |  | [View](https://fastrouter.ai/models/deepseek/deepseek-v3.2) |
| [Fireworks AI](/llm/fireworks.txt) | Deepseek v3.2 | 0.56 | 1.68 |  |  |
| [Helicone](/llm/helicone.txt) | DeepSeek V3.2 | 0.26 | 0.40 |  | [View](https://www.helicone.ai/model/deepseek-v3.2) |
| [Mammouth AI](/llm/mammouth.txt) | deepseek-v3.2 | 0.30 | 0.45 |  |  |
| [Abacus](/llm/abacus.txt) | DeepSeek V3.2 | 0.27 | 0.40 |  |  |
| [Novita AI](/llm/novita.txt) | deepseek-v3.2 | 0.27 | 0.40 |  |  |
| [Azure OpenAI](/llm/azure.txt) | DeepSeek-V3.2 | 0.58 | 1.68 |  |  |
| [Baseten](/llm/baseten.txt) | DeepSeek V3.2 | 0.30 | 0.45 |  |  |
| [SiliconFlow](/llm/siliconflow.txt) | DeepSeek-V3.2 | 0.27 | 0.42 |  | [View](https://www.siliconflow.com./models/deepseek-v3-2) |
| [Hugging Face](/llm/huggingface.txt) | DeepSeek-V3.2 | 0.28 | 0.40 |  |  |
| [iFlow](/llm/iflowcn.txt) | DeepSeek-V3.2-Exp | 0.00 | 0.00 | Yes |  |
| [Azure AI Services](/llm/azurecognitiveservices.txt) | DeepSeek-V3.2 | 0.58 | 1.68 |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | DeepSeek V3.2 |  |  |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | DeepSeek V3.2 Thinking |  |  |  |  |
| [Nano-GPT](/llm/nanogpt.txt) | DeepSeek V3.2 TEE |  |  |  |  |
| [Ollama Cloud](/llm/ollama.txt) | deepseek-v3.2 |  |  |  | [View](https://ollama.com/library/deepseek-v3.2) |
| [OpenRouter](/llm/openrouter.txt) | DeepSeek V3.2 | 0.25 | 0.40 |  | [View](https://openrouter.ai/deepseek/deepseek-v3.2-20251201) |
| [Poe](/llm/poe.txt) | DeepSeek-V3.2 | 0.27 | 0.40 |  | [View](https://poe.com/deepseek-v3.2/api) |
| [Synthetic.new](/llm/synthetic.txt) | deepseek-ai/DeepSeek-V3.2 | 0.56 | 1.68 |  |  |
| [ValorGPT](/llm/valorgpt.txt) | DeepSeek V3.2 |  |  |  | [View](https://www.valorgpt.com/models/deepseek-deepseek-v3.2) |
| [Venice](/llm/venice.txt) | DeepSeek V3.2 | 0.40 | 1.00 |  |  |
| [Vercel AI Gateway](/llm/vercel.txt) | DeepSeek V3.2 | 0.26 | 0.38 |  |  |
| [Vivgrid](/llm/vivgrid.txt) |  | 0.28 | 0.42 |  |  |
| [Yupp](/llm/yupp.txt) | DeepSeek V3.2 Chat (OpenRouter) |  |  |  |  |
| [Yupp](/llm/yupp.txt) | DeepSeek V3.2 (Sambanova) |  |  |  |  |
| [ZenMUX](/llm/zenmux.txt) | DeepSeek: DeepSeek V3.2 | 0.28 | 0.43 |  |  |
| [Atlas Cloud](/llm/atlascloud.txt) | DeepSeek V3.2 | 0.26 | 0.38 |  | [View](https://www.atlascloud.ai/models/deepseek-ai/deepseek-v3.2) |
| [DeepInfra](/llm/deepinfra.txt) | DeepSeek-V3.2 | 0.26 | 0.38 |  |  |
| [Nebius Token Factory](/llm/nebius.txt) | DeepSeek-V3.2 | 0.30 | 0.45 |  | [View](https://huggingface.co/deepseek-ai/DeepSeek-V3.2) |
| [Routeway](/llm/routeway.txt) | DeepSeek: DeepSeek V3.2 | 0.25 | 0.39 |  | [View](https://routeway.ai/models) |
| [Nvidia](/llm/nvidia.txt) | deepseek-v3.2 | 0.00 | 0.00 | Yes | [View](https://build.nvidia.com/deepseek-ai/deepseek-v3.2) |
| [Yupp](/llm/yupp.txt) | DeepSeek V3.2 Chat (Baseten) |  |  |  |  |
| [302.AI](/llm/302ai.txt) | deepseek-v3.2 | 0.29 | 0.43 |  | [View](https://302ai-en.apifox.cn/api-207705121) |
| [Kilo Code](/llm/kilocode.txt) | DeepSeek: DeepSeek V3.2 | 0.25 | 0.38 |  |  |
| [GMI Cloud](/llm/gmi.txt) | DeepSeek-V3.2 | 0.28 | 0.40 |  |  |
| [Parasail](/llm/parasail.txt) | Deepseek V32 | 0.28 | 0.45 |  | [View](https://www.saas.parasail.io/pricing) |
| [RedPill](/llm/redpill.txt) | DeepSeek: DeepSeek V3.2 | 0.27 | 0.40 |  |  |
| [SambaNova AI](/llm/sambanova.txt) | DeepSeek-V3.2 | 3.00 | 4.50 |  |  |
| [IO.NET](/llm/ionet.txt) | DeepSeek-V3.2 | 0.25 | 0.38 |  |  |
| [Google Vertex AI](/llm/googlevertex.txt) | Deepseek V32 |  |  |  |  |
| [Baidu AI Studio](/llm/baidu.txt) |  |  |  |  | [View](https://aistudio.baidu.com) |

---

[← Back to all providers](/llm.txt)