DeepSeek V3.2-Exp

Name: DeepSeek-V3.2-Exp
Author: DeepSeek-AI

Reasoning modelDeepSeek VDeepSeek V3.2

DeepSeek-V3.2-Exp

Release date: 2025-09-29Updated: 2026-06-15 07:18:20.6821,900

Live demo GitHub Hugging Face Compare

Parameters

671B

Context length

128K

Chinese support

Not supported

Reasoning ability

DeepSeek V3.2-Exp 是 DeepSeek 于2025年9月发布的实验版本，首次引入 DeepSeek Sparse Attention（DSA）稀疏注意力机制，在保持与 V3.1-Terminus 相当性能的同时，将长上下文推理速度提升 2–3 倍，API 定价下调超过 50%。

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

DeepSeek V3.2-Exp

Model basics

Reasoning traces

Supported

Thinking modes

Thinking Mode (Default)

Context length

128K tokens

Max output length

64K tokens

Model type

Reasoning model

Modality (in / out)

Text → Text

Release date

2025-09-29

Model file size

1342GB

MoE architecture

Yes

Total params / Active params

671B / 37B

Knowledge cutoff

No data

DeepSeek V3.2-Exp

Open source & experience

Code license

MIT License

Weights license

MIT License- 免费商用授权

GitHub repo

https://github.com/deepseek-ai/DeepSeek-V3.2-Exp

Hugging Face

https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Exp

Live demo

https://chat.deepseek.com

DeepSeek V3.2-Exp

Official resources

Paper

DeepSeek-V3.2-Exp: Boosting Long-Context Efficiency with DeepSeek Sparse Attention

DataLearnerAI blog

No blog post yet

DeepSeek V3.2-Exp

API details

API speed

3/5

No public API pricing yet.

DeepSeek V3.2-Exp

Benchmark Results

DeepSeek V3.2-Exp currently shows benchmark results led by SimpleQA (1 / 45, score 97.10), Aider-Polyglot (11 / 59, score 74.20), MMLU Pro (25 / 126, score 85). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.