LIVE INTELLIGENCE · part of the Xentir AI Index

DeepSeek-V3.2 (Reasoner) vs o4-mini

Name: DeepSeek-V3.2 (Reasoner) vs o4-mini — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

DeepSeek-V3.2 (Reasoner) OPEN WEIGHTS · DeepSeek, released 2025-12-01
o4-mini · OpenAI, released 2025-04-16

“Epoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks' [online resource].” · epoch.ai/benchmarks · Licensed CC BY 4.0 · data last verified 2026-09-12 · Methodology · Data Sources

DeepSeek-V3.2 (Reasoner) leads on more of the 4 benchmarks where both are measured — DeepSeek-V3.2 (Reasoner) on 3, o4-mini on 1. The remaining 5 of 9 tracked benchmarks cannot be compared: 4 have a result for only one of the two models, 1 for neither. Nothing on this page is estimated to fill those in.

Head-to-head

Measured on 4 of 9 tracked benchmarks for both models — DeepSeek-V3.2 (Reasoner) leads on 3, o4-mini leads on 1.

Benchmark	DeepSeek-V3.2 (Reasoner)	o4-mini	Leads
Epoch Capabilities Index (ECI)	146.2	145.7	DeepSeek-V3.2 (Reasoner)
SWE-bench Verified	Not measured	Not measured	Not measured on both
GPQA Diamond	0.834	0.796 (high)	DeepSeek-V3.2 (Reasoner)
MATH Level 5	Not measured	0.978 (high)	Comparison unavailable — DeepSeek-V3.2 (Reasoner) has no published result
OTIS Mock AIME 2024-2025	0.878	0.817 (high)	DeepSeek-V3.2 (Reasoner)
FrontierMath	Not measured	0.248 (high)	Comparison unavailable — DeepSeek-V3.2 (Reasoner) has no published result
FrontierMath Tier 4	Not measured	0.062 (high)	Comparison unavailable — DeepSeek-V3.2 (Reasoner) has no published result
SimpleQA Verified	Not measured	0.196 (low)	Comparison unavailable — DeepSeek-V3.2 (Reasoner) has no published result
Chess Puzzles	0.140	0.260 (high)	o4-mini

Benchmark results measure specific tasks and should not be treated as a universal measure of model quality. A model with no row here for a benchmark has not been measured on it — that is left blank, never estimated.

What remains uncertain

DeepSeek-V3.2 (Reasoner) has no published result on FrontierMath, FrontierMath Tier 4, MATH Level 5, SimpleQA Verified.
1 tracked benchmark has no published result for either model.

A missing result means Xentir has not found one published by the source it cites, not that the model performs badly. See Data Sources.

Reference price

Per 1M tokens	DeepSeek-V3.2 (Reasoner)	o4-mini
Output	—	$4.40

List price as published by the vendor. Xentir does not verify negotiated or enterprise rates, and price is not a measure of capability — see Data Sources.

Full profiles

DeepSeek-V3.2 (Reasoner) → · o4-mini → · the full AI Model Rankings table