DeepSeek-V3.2-Speciale

Name: DeepSeek-V3.2-Speciale
Author: DeepSeek

DeepSeek

DeepSeek-V3.2-Speciale is a specialized version of DeepSeek-V3.2, optimized for enhanced performance on specific tasks. A 685B parameter model showing excellent results in math and coding.

Key Specifications

Parameters

685.0B

Context

131.1K

Release Date

November 30, 2025

Average Score

70.7%

API Documentation Research Paper

Timeline

Key dates in the model's history

Announcement

November 30, 2025

Last Update

January 22, 2026

Today

March 25, 2026

Technical Specifications

Parameters

685.0B

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)

$0.28

Output (per 1M tokens)

$0.42

Max Input Tokens

131.1K

Max Output Tokens

131.1K

Supported Features

Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Programming

Programming skills tests

SWE-Bench Verified

Resolved • Self-reported

73.0%

Other Tests

Specialized benchmarks

HMMT 2025

Pass@1 • Self-reported

99.0%

AIME 2025

Pass@1 • Self-reported

96.0%

CodeForces

Standard evaluation • Self-reported

30.0%

T2-Bench

Pass@1 • Self-reported

80.0%

Terminal-Bench 2.0

Accuracy • Self-reported

46.0%

License & Metadata

License

mit

Announcement Date

November 30, 2025

Last Updated

January 22, 2026

Similar Models

All Models

DeepSeek-V2.5

DeepSeek

236.0B

Best score:0.9 (HumanEval)

Released:May 2024

Price:$2.00/1M tokens

DeepSeek-V3.2 (Non-thinking)

DeepSeek

685.0B

Released:Nov 2025

Price:$0.28/1M tokens

DeepSeek

685.0B

Best score:0.8 (GPQA)

Released:Sep 2025

Price:$0.27/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.