ModelBeat
Models/DeepSeek-V3.2
All models
D

DeepSeek: DeepSeek-V3.2

deepseek/deepseek-v3.2

DeepSeek-V3.2 is DeepSeek’s open-weights AI model, released Dec 1, 2025. It costs $0.25/M input and $0.42/M output tokens, and handles a 164K-token context window.

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.

Context
164K
Input
$0.25 / 1M
Output
$0.42 / 1M
Released
Dec 1, 2025
Modalities
texttext

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where DeepSeek-V3.2 lands among all models tracked on Model Beat.

38.0
Intelligence Index
Epoch AI
38th percentile of tracked models
42.0
Coding Index
Epoch AI
42nd percentile of tracked models
39.0
Agentic Index
Epoch AI
39th percentile of tracked models

Reasoning

8 evals
ARC-AGI57.0%

Abstract visual reasoning

ARC-AGI-24.0%

Harder abstract reasoning

GPQA Diamond83.4%

Graduate-level scientific reasoning

Humanity's Last Exam22.2%

Frontier of human expert knowledge

MMLU-Pro86.2%

Broad expert knowledge

SimpleBench52.6%

Common-sense trick questions

SimpleQA Verified27.5%

Factual accuracy & hallucination

WeirdML46.7%

Novel ML problem-solving

Coding

3 evals
LiveCodeBench86.2%

Contamination-free coding

SciCode38.9%

Scientific research coding

WebDev Arena1286

Human-rated web development

Math

3 evals
AIME 2024/202587.8%

Olympiad-qualifier math

FrontierMath22.1%

Research-level math problems

FrontierMath Tier 42.1%

Hardest research math

Agentic & Tools

3 evals
APEX7.0%

Multi-step agentic tasks

Terminal-Bench39.6%

Command-line agentic tasks

τ²-bench90.6%

Tool-agent-user reliability

In the news

1

Changelog

6
  • DeepSeek-V3.2: cheapest credible provider via SiliconFlow now $0.42 per 1M output tokens (was $0.378, +11%)cheapest provider · Jul 29, 2026
  • DeepSeek-V3.2: cheapest credible provider via Baidu now $0.252 per 1M input tokens (was $0.2145, +17%)cheapest provider · Jul 16, 2026
  • DeepSeek-V3.2: cheapest credible provider via Baidu now $0.378 per 1M output tokens (was $0.3217, +18%)cheapest provider · Jul 16, 2026
  • DeepSeek-V3.2: context window changed from 131,072 to 163,840 tokenscontext · Jul 15, 2026
  • DeepSeek-V3.2: cheapest credible provider via StreamLake now $0.2145 per 1M input tokens (was $0.2288, -6%)cheapest provider · Jul 11, 2026
  • DeepSeek-V3.2: cheapest credible provider via StreamLake now $0.3217 per 1M output tokens (was $0.3432, -6%)cheapest provider · Jul 11, 2026

Compare DeepSeek-V3.2 with

Frequently asked questions

What is 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.

Who created DeepSeek-V3.2?

DeepSeek-V3.2 is an AI model developed by DeepSeek, part of the DeepSeek family, released Dec 1, 2025.

How much does DeepSeek-V3.2 cost?

DeepSeek-V3.2 is served from $0.25 per million input tokens and $0.42 per million output tokens (cheapest credible provider on OpenRouter: SiliconFlow).

What is DeepSeek-V3.2's context window?

DeepSeek-V3.2 supports a context window of up to 164K tokens.

How does DeepSeek-V3.2 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, DeepSeek-V3.2 scores 86.2% on LiveCodeBench, 86.2% on MMLU-Pro, 90.6% on τ²-bench.

Is DeepSeek-V3.2 open source?

DeepSeek-V3.2 is released with open weights (Open weights (unrestricted)), so the model can be downloaded and self-hosted.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Jul 30, 2026.