Question 1

What is the cost of DeepSeek-V3.2 by Deep Infra?

Accepted Answer

DeepSeek-V3.2 by Deep Infra costs $0.26 per 1M input tokens and $0.38 per 1M output tokens. Cached reads cost $0.13 per 1M tokens.

Question 2

What is the context window of DeepSeek-V3.2 by Deep Infra?

Accepted Answer

DeepSeek-V3.2 by Deep Infra has a context window of 164K tokens. It supports up to 164K input tokens and can generate up to 64K output tokens.

Question 3

What are the capabilities of DeepSeek-V3.2 by Deep Infra?

Accepted Answer

DeepSeek-V3.2 by Deep Infra supports advanced reasoning, tool calling/function calling, structured output, adjustable temperature. It has closed weights.

Question 4

What input and output types does DeepSeek-V3.2 support?

Accepted Answer

DeepSeek-V3.2 by Deep Infra accepts text as input and can generate text as output.

Question 5

Does DeepSeek-V3.2 by Deep Infra support reasoning?

Accepted Answer

Yes, DeepSeek-V3.2 by Deep Infra supports advanced reasoning capabilities.

Question 6

Is DeepSeek-V3.2 by Deep Infra open source?

Accepted Answer

No, DeepSeek-V3.2 by Deep Infra has closed weights. The model is only available through the Deep Infra API.

Question 7

Does DeepSeek-V3.2 support function calling?

Accepted Answer

Yes, DeepSeek-V3.2 by Deep Infra supports tool calling (also known as function calling), allowing it to interact with external tools and APIs during conversations.

Question 8

What is the knowledge cutoff date for DeepSeek-V3.2?

Accepted Answer

DeepSeek-V3.2 by Deep Infra has a knowledge cutoff date of 2024-12. This means the model was trained on data available up until that date.

DeepSeek-V3.2 API

Input Modalities

Output Modalities

Standard (per 1M tokens)

Caching (per 1M tokens)