ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

GPT-5.4 ProvsStep 3.7 Flash

GPT-5.4 Pro leads 1 of 1 shared benchmark. 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
1 – 0
GPT-5.4 Pro leads
Cheaper per token
list price, input + output
Larger context
GPT-5.4 Pro
1.1M tokens
Provider uptime (30d)
95.2941% · —
OpenAI · StepFun
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
Step 3.7 Flash
StepFun · released 2026-05-29
textvision
Context
256K
Max out
Input /1M
Output /1M
Cached /1M
Scores
19 · 6 core
Quality

Benchmark matrix

BenchmarkGPT-5.4 ProStep 3.7 FlashΔEdge
Reported by both · 1
BrowseComp89.3% 75.8% +13.5 ptGPT-5.4 Pro
Only GPT-5.4 Pro reports · 11
GPQA Diamond94.4% not reported
Humanity's Last Exam (no tools)42.7% not reported
ARC-AGI-1 (Verified)94.5% not reported
ARC-AGI-2 (Verified)83.3% not reported
FrontierMath Tier 1-350% not reported
FrontierMath Tier 438% not reported
FrontierMath Tier 4 v2 (Epoch AI run)58.5% epochnot reported
FrontierMath Tiers 1-3 v2 (Epoch AI run)82.5% epochnot reported
GDPval (wins or ties)82% not reported
Humanity's Last Exam (with tools)58.7% not reported
SimpleQA Verified46.3% epochnot reported
Only Step 3.7 Flash reports · 18
SWE-bench Verifiednot reported76.5%
AIME 2026not reported95% matharena
AA-LCRnot reported63.9%
ClawEval v1.1not reported67.1%
DeepSearchQA Accuracynot reported81.7%
DeepSearchQA F1not reported92.8%
GDPval-Stirrupnot reported1415.8
GDPval-Stirrup Intelligence Indexnot reported45.8
HLE (with tools, text-only)not reported49.7%
HLE (with tools)not reported47.2%
ResearchRubricsnot reported71.7%
SimpleVQA (with tools)not reported79.2%
SWE-bench Multilingualnot reported72.4%
SWE-bench Pronot reported56.3%
Terminal-Bench 2.1not reported59.6%
Toolathlonnot reported49.5%
V* (with Python)not reported95.3%
WorldVQA (with visual search)not reported58.1%
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-5.4 Pro minus Step 3.7 Flash 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-5.4 ProStep 3.7 FlashEdge
Context window1.1M tokens256K tokensGPT-5.4 Pro
Max output128K tokens
Input price / 1M$30
Output price / 1M$180
Cached input / 1M
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$210
Modalitiestext · visiontext · visionTie
Released2026-03-052026-05-29
Cited benchmark scores1319
Reliability

Provider status

All providers
More matchups

GPT-5.4 Pro vs …

More matchups

Step 3.7 Flash vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run GPT-5.4 Pro and Step 3.7 Flash on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.