# DeepSeek-V3.2

> DeepSeek-V3.2 is an open-weight language model from DeepSeek, released on 2 Dec 2025. It accepts text and generates text, with a 164K-token context window and up to 64K output tokens per response. As of 4 Oct 2026, the lowest listed API price is $0.26 per million input tokens and $0.38 per million output tokens (DeepInfra), across 9 providers we track.

Source page: https://www.aimodel.directory/models/deepseek-v3-2
Last verified: 4 Oct 2026

## Specifications

- **Developer:** DeepSeek
- **Release date:** 2 Dec 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Weights:** Open weights (source: Hugging Face, https://huggingface.co/deepseek-ai/DeepSeek-V3.2)
- **License:** mit (source: Hugging Face, https://huggingface.co/deepseek-ai/DeepSeek-V3.2)
- **Parameters:** 685B (source: Hugging Face, https://huggingface.co/deepseek-ai/DeepSeek-V3.2)
- **Context window:** 164K (163,840 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Max output:** 64K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Knowledge cutoff:** Dec 2024 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Reasoning mode:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/deepinfra/models/deepseek-ai/DeepSeek-V3.2.toml)
- **Hugging Face:** deepseek-ai/DeepSeek-V3.2 (source: Hugging Face, https://huggingface.co/deepseek-ai/DeepSeek-V3.2)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| DeepInfra | $0.26 | $0.38 | $0.13 | 164K |
| Novita AI | $0.27 | $0.40 | $0.13 | 164K |
| SiliconFlow | $0.27 | $0.42 | $0.14 | 164K |
| Hugging Face Inference Providers | $0.28 | $0.40 | – | 164K |
| OpenRouter | $0.28 | $0.42 | $0.03 | 131K |
| FriendliAI | $0.50 | $1.50 | $0.25 | 164K |
| Azure AI Foundry | $0.58 | $1.68 | – | 128K |
| Amazon Bedrock | $0.62 | $1.85 | – | 164K |
| Vercel AI Gateway | $0.62 | $1.85 | – | 128K |

## FAQ

### What is the context window of DeepSeek-V3.2?

DeepSeek-V3.2 has a context window of 163,840 tokens (164K) and can generate up to 64,000 tokens in a single response, according to models.dev.

### How much does the DeepSeek-V3.2 API cost?

DeepSeek-V3.2 is listed from $0.26 per million input tokens and $0.38 per million output tokens (DeepInfra). Prices last verified 4 Oct 2026.

### Is DeepSeek-V3.2 open source?

DeepSeek-V3.2 is an open-weight model: its weights are published on Hugging Face as deepseek-ai/DeepSeek-V3.2 under the mit license. Check the license terms before commercial use.

### When was DeepSeek-V3.2 released?

DeepSeek released DeepSeek-V3.2 on 2 Dec 2025, according to models.dev.

### Which providers offer DeepSeek-V3.2?

We track DeepSeek-V3.2 on 9 providers: DeepInfra, Novita AI, SiliconFlow, Hugging Face Inference Providers, OpenRouter, FriendliAI, Azure AI Foundry, Amazon Bedrock, Vercel AI Gateway.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "DeepSeek-V3.2", https://www.aimodel.directory/models/deepseek-v3-2