VENICE ATLAS · MODEL DOSSIER · DEEPSEEK V3.2

Model overview

DeepSeek V3.2

Name: DeepSeek V3.2
Brand: venice.ai
Price: 0.3300 USD
Availability: InStock

DeepSeek-V3.2 is an efficient large language model with DeepSeek Sparse Attention (DSA) for long contexts. It features strong reasoning and tool-use skills, achieving top results on the 2025 IMO and IOI.

deepseek-v3.2 · venice.ai

Text · Venice catalog$0.33 per 1M tokensPrice on Venice AI

ModalityTextOn Venice AI

Primary price$0.33 per 1M tokensper 1M tokens

Context window160KMax tokens

Benchmarks17Benchmark results

Benchmark results

See the tests behind each score. They use the benchmark’s original scoring scale.

Finance#1926.33 rating Language#2026.33 rating Legal#1426.33 rating

Agents4 results

MCP-Markmcp-mark · agents

38.0%

ReportedOct 2026

MCP-Universemcp-universe · agents

45.9%

ReportedOct 2026

t2-bencht2-bench · agents

80.3%

ReportedOct 2026

Toolathlontoolathlon · agents

35.2%

ReportedOct 2026

Language1 result

MMLU-Prommlu-pro · language

85.0%

ReportedOct 2026

Math4 results

AIME 2025aime-2025 · math

93.1%

ReportedOct 2026

CodeForcescodeforces · math

79.5%

ReportedOct 2026

HMMT 2025hmmt-2025 · math

90.2%

ReportedOct 2026

IMO-AnswerBenchimo-answerbench · math

78.3%

ReportedOct 2026

Reasoning8 results

BrowseCompbrowsecomp · reasoning

51.4%

ReportedOct 2026

BrowseComp-zhbrowsecomp-zh · reasoning

65.0%

ReportedOct 2026

GPQAgpqa · reasoning

82.4%

ReportedOct 2026

Humanity's Last Examhumanity's-last-exam · reasoning

40.8%

ReportedOct 2026

LiveCodeBenchlivecodebench · reasoning

83.3%

ReportedOct 2026

SWE-bench Multilingualswe-bench-multilingual · reasoning

70.2%

ReportedOct 2026

SWE-Bench Verifiedswe-bench-verified · reasoning

73.1%

ReportedOct 2026

Terminal-Bench 2.0terminal-bench-2 · reasoning

46.4%

ReportedOct 2026

Learn what each benchmark measures ↗

Pricing detail

Input price per 1M tokens: $0.33
Output price per 1M tokens: $0.48
Cache read price per 1M tokens: $0.16
Blended price per 1M tokens (50/50): $0.41

Prices are published by Venice AI. Venice model page ↗

Capabilities

ReasoningFunction callingWeb search

Endpoint metadata

Privacy: private
Quantization: not-available
Maximum completion tokens: 33K
Uncensored: No
Beta: No
Mapping confidence: 100.0%

View model source ↗

All capabilities

Quantization: not-available
Supports E2 E E: No
Supports Vision: No
Supports X Search: No
Optimized For Code: No
Supports Log Probs: No
Supports Reasoning: Yes
Supports Web Search: Yes
Supports Audio Input: No
Supports Video Input: No
Default Reasoning Effort: low
Reasoning Effort Options: none, low, medium, high
Supports Multiple Images: No
Supports Response Schema: Yes
Supports Tee Attestation: No
Supports Function Calling: Yes
Supports Reasoning Effort: Yes

Related models

More from deepseek

deepseek-v4-1-flash-1017DeepSeek v4.1 Flash (1017)$0.30 input per 1M tokens ↗deepseek-v4-pro-0813DeepSeek V4 Pro 0813$1.65 input per 1M tokens ↗deepseek-v4-1-flashDeepSeek V4.1 Flash$0.30 input per 1M tokens ↗