ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GPT-5.2vsGPT-5.4 Pro

GPT-5.4 Pro leads 7 of 7 shared benchmarks. GPT-5.2 is 13.3x cheaper per token. GPT-5.4 Pro has the larger context window (1.1M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
0 – 7
GPT-5.4 Pro leads
Cheaper per token
GPT-5.2
13.3x cheaper, input + output
Larger context
GPT-5.4 Pro
1.1M tokens
Provider uptime (30d)
95.389%
OpenAI
GPT-5.2
OpenAI · released 2025-12-11
textvision
Context
400K
Max out
128K
Input /1M
$1.75
Output /1M
$14
Cached /1M
$0.175
Scores
16 · 11 core
GPT-5.4 Pro
OpenAI · released 2026-03-05
textvision
Context
1.1M
Max out
128K
Input /1M
$30
Output /1M
$180
Cached /1M
Scores
13 · 8 core
Quality

Benchmark matrix

BenchmarkGPT-5.2GPT-5.4 ProΔEdge
Reported by both · 7
GPQA Diamond92.4% 94.4% -2 ptGPT-5.4 Pro
ARC-AGI-1 (Verified)86.2% 94.5% -8.3 ptGPT-5.4 Pro
ARC-AGI-2 (Verified)52.9% 83.3% -30.4 ptGPT-5.4 Pro
FrontierMath Tier 4 v2 (Epoch AI run)31.7% epoch58.5% epoch-26.8 ptGPT-5.4 Pro
FrontierMath Tiers 1-3 v2 (Epoch AI run)67.4% epoch82.5% epoch-15.1 ptGPT-5.4 Pro
GDPval (wins or ties)70.9% 82% -11.1 ptGPT-5.4 Pro
SimpleQA Verified37.1% epoch46.3% epoch-9.2 ptGPT-5.4 Pro
Only GPT-5.2 reports · 7
SWE-bench Verified80% not reported
AIME 2025100% not reported
AIME 202698.3% matharenanot reported
LMArena Elo1412.4 not reported
OTIS Mock AIME 2024-2025 (Epoch AI run)96.1% epochnot reported
SWE-Bench Pro55.6% not reported
SWE-bench Verified (Epoch AI run)73.8% epochnot reported
Only GPT-5.4 Pro reports · 5
Humanity's Last Exam (no tools)not reported42.7%
BrowseCompnot reported89.3%
FrontierMath Tier 1-3not reported50%
FrontierMath Tier 4not reported38%
Humanity's Last Exam (with tools)not reported58.7%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GPT-5.2 minus GPT-5.4 Pro in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecGPT-5.2GPT-5.4 ProEdge
Context window400K tokens1.1M tokensGPT-5.4 Pro
Max output128K tokens128K tokensTie
Input price / 1M$1.75 $30 GPT-5.2
Output price / 1M$14 $180 GPT-5.2
Cached input / 1M$0.175
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$15.75$210GPT-5.2
Modalitiestext · visiontext · visionTie
Released2025-12-112026-03-05
Cited benchmark scores1613
Reliability

OpenAI status

All providers
More matchups

GPT-5.4 Pro vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GPT-5.2 and GPT-5.4 Pro on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.