LIVE INTELLIGENCE · part of the Xentir AI Index

DeepSeek-V4-Pro vs GPT-5.4 Mini

Name: DeepSeek-V4-Pro vs GPT-5.4 Mini — benchmark results
Creator: Epoch AI
License: https://creativecommons.org/licenses/by/4.0/

DeepSeek-V4-Pro OPEN WEIGHTS · DeepSeek, released 2026-04-24
GPT-5.4 Mini · OpenAI, released 2026-03-17

“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 2026-09-09 · Methodology · Data Sources

DeepSeek-V4-Pro leads on more of the 5 benchmarks where both are measured — DeepSeek-V4-Pro on 4, GPT-5.4 Mini on 1. The remaining 4 of 9 tracked benchmarks cannot be compared: 3 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 5 of 9 tracked benchmarks for both models — DeepSeek-V4-Pro leads on 4, GPT-5.4 Mini leads on 1.

Benchmark	DeepSeek-V4-Pro	GPT-5.4 Mini	Leads
Epoch Capabilities Index (ECI)	149.1	149.0	DeepSeek-V4-Pro
SWE-bench Verified	0.776 (max)	Not measured	Comparison unavailable — GPT-5.4 Mini has no published result
GPQA Diamond	0.909 (high)	0.869 (xhigh)	DeepSeek-V4-Pro
MATH Level 5	Not measured	Not measured	Not measured on both
OTIS Mock AIME 2024-2025	0.967 (max)	0.889 (xhigh)	DeepSeek-V4-Pro
FrontierMath	Not measured	0.283 (high)	Comparison unavailable — DeepSeek-V4-Pro has no published result
FrontierMath Tier 4	Not measured	0.021 (high)	Comparison unavailable — DeepSeek-V4-Pro has no published result
SimpleQA Verified	0.470 (max)	0.294 (high)	DeepSeek-V4-Pro
Chess Puzzles	0.200 (max)	0.240 (xhigh)	GPT-5.4 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-V4-Pro has no published result on FrontierMath, FrontierMath Tier 4.
GPT-5.4 Mini has no published result on SWE-bench 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-V4-Pro	GPT-5.4 Mini
Output	—	$4.50

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-V4-Pro → · GPT-5.4 Mini → · the full AI Model Rankings table