Skip to main content

Model comparison

GPT-5.4 mini vs GPT-5.5

Data verified

Head-to-head evidence from 30 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

56.77/100
Margin
16.7pts
winning →
OpenAI
73.51/100
1 category wins3 category wins

Public leaderboard positions: GPT-5.4 mini #75 (Estimated); GPT-5.5 #9 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. GPT-5.4 mini and GPT-5.5 share 30 comparable benchmark results. 4 of 8 categories are comparable. 0 results are unique to GPT-5.4 mini; 27 to GPT-5.5.

Updated July 22, 2026
Shared results
30
GPT-5.4 mini only
0
GPT-5.5 only
27
Comparable categories
4 / 8

Pick GPT-5.5 if you want the stronger benchmark profile. GPT-5.4 mini only becomes the better choice if multimodal & grounded is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 30 shared benchmark results across 7 evidence categories; 4 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

GPT-5.5 is clearly ahead on the BenchAlign aggregate, 73.51 to 56.77. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

GPT-5.5's sharpest advantage is in mathematics, where it averages 47.6 against 21.7. The single biggest benchmark swing on the page is FrontierMath v2 (Tier 4), 2.080% to 35.400%. GPT-5.4 mini does hit back in multimodal & grounded, so the answer changes if that is the part of the workload you care about most.

GPT-5.5 is also the more expensive model on tokens at $5.00 input / $30.00 output per 1M tokens, versus $0.75 input / $4.50 output per 1M tokens for GPT-5.4 mini. That is roughly 6.7x on output cost alone. GPT-5.5 gives you the larger context window at 1M, compared with 400K for GPT-5.4 mini.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for GPT-5.4 mini and GPT-5.5
CategoryGPT-5.4 miniΔGPT-5.5
MathGPT-5.4 mini21.7Margin 25.9GPT-5.547.6
AgenticGPT-5.4 mini65.7Margin 15.9GPT-5.581.6
KnowledgeGPT-5.4 mini47.8Margin 10.0GPT-5.557.8
MultimodalGPT-5.4 mini76.6Margin 6.2GPT-5.570.4
CodingGPT-5.4 miniNot measuredMarginNo overlapGPT-5.558.6
ReasoningGPT-5.4 miniNot measuredMarginNo overlapGPT-5.585.0

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · GPT-5.4 miniB · GPT-5.5
  1. FrontierMath v2 (Tier 4)

    Math
    Source ↗
    A 2.080%B 35.400%
    Winner: GPT-5.5Δ 33.3
    FrontierMath v2 (Tier 4): GPT-5.4 mini scored 2.080%; GPT-5.5 scored 35.400%. GPT-5.5 wins this benchmark.
  2. FrontierMath v2 (Tiers 1-3)

    Math
    Source ↗
    A 28.280%B 51.700%
    Winner: GPT-5.5Δ 23.4
    FrontierMath v2 (Tiers 1-3): GPT-5.4 mini scored 28.280%; GPT-5.5 scored 51.700%. GPT-5.5 wins this benchmark.
  3. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 60%B 82%
    Winner: GPT-5.5Δ 22
    Terminal-Bench 2.0: GPT-5.4 mini scored 60%; GPT-5.5 scored 82%. GPT-5.5 wins this benchmark.
  4. HLE

    Knowledge
    Source ↗
    A 41.5%B 52.2%
    Winner: GPT-5.5Δ 10.7
    HLE: GPT-5.4 mini scored 41.5%; GPT-5.5 scored 52.2%. GPT-5.5 wins this benchmark.
  5. OSWorld-Verified

    Agentic
    Source ↗
    A 72.1%B 78.7%
    Winner: GPT-5.5Δ 6.6
    OSWorld-Verified: GPT-5.4 mini scored 72.1%; GPT-5.5 scored 78.7%. GPT-5.5 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricGPT-5.4 miniGPT-5.5Comparison
Input / output priceUSD per 1M tokensGPT-5.4 mini$0.75 input / $4.5 outputGPT-5.5$5 input / $30 outputGPT-5.4 mini has the lower combined listed price.
Generation speedtokens per secondGPT-5.4 mini201 tok/sGPT-5.5Not availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenGPT-5.4 mini3.85 sGPT-5.5Not availableA complete latency comparison is not available.
Context windowmaximum listed tokensGPT-5.4 mini400KGPT-5.51MGPT-5.5 lists the larger context window.

Benchmark Deep Dive

AgenticGPT-5.5 wins
BenchmarkGPT-5.4 miniGPT-5.5Result
Terminal-Bench 2.0Source 60%82%GPT-5.5 leads
OSWorld-VerifiedSource 72.1%78.7%GPT-5.5 leads
MCP AtlasSource 57.7%75.3%GPT-5.5 leads
ToolathlonSource 42.9%55.6%GPT-5.5 leads
τ²-bench resultsSource 83.3%93.9%GPT-5.5 leads
AA Agentic IndexSource 30.2%44.9%GPT-5.5 leads
APEX-Agents-AASource 28.2%37.7%GPT-5.5 leads
GDPval-AASource 33.6%49.5%GPT-5.5 leads
GDPval-AASource 11711490GPT-5.5 leads
CyberGymSource 81.8%Not comparable
BrowseCompSource 84.4%Not comparable
Gert LabsSource 72.93%Not comparable
ResearchClawBenchSource 17.0%Not comparable
OSWorld 2.0Source 13.0%Not comparable
JobBenchSource 42.7%Not comparable
ExploitGymSource 13.4%Not comparable
AA BriefcaseSource 1154Not comparable
AA AutomationBenchSource 42.1%Not comparable
AA EnterpriseOps-GymSource 46.6%Not comparable
AA Harvey LABSource 86.3%Not comparable
AA ITBenchSource 45.8%Not comparable
AA Tau3 BankingSource 31.3%Not comparable
terminalBenchHardSource 60.6%Not comparable
aaTerminalBench21Source 84.3%Not comparable
Coding
BenchmarkGPT-5.4 miniGPT-5.5Result
Vibe Code BenchSource 47.97%69.85%GPT-5.5 leads
AA Coding IndexSource 56.1%74.9%GPT-5.5 leads
AA-SciCodeSource 49.9%56.1%GPT-5.5 leads
FrontierCode 1.1 MainSource 27.0%43.0%GPT-5.5 leads
SWE-bench ProSource 58.6%Not comparable
Terminal-Bench 2.0Source 82.0%Not comparable
React Native EvalsSource 84.7%Not comparable
cursorBench31Source 59.2%Not comparable
cursorBench32Source 58.4%Not comparable
Reasoning
BenchmarkGPT-5.4 miniGPT-5.5Result
AA-LCRSource 69.3%74.3%GPT-5.5 leads
CritPtSource 10.0%27.1%GPT-5.5 leads
MRCR v2 64K-128KSource 83.1%Not comparable
MRCR v2 128K-256KSource 87.5%Not comparable
ARC-AGI-2Source 85%Not comparable
KnowledgeGPT-5.5 wins
BenchmarkGPT-5.4 miniGPT-5.5Result
GPQASource 88%93.6%GPT-5.5 leads
HLESource 41.5%52.2%GPT-5.5 leads
HLE w/o toolsSource 28.2%41.4%GPT-5.5 leads
Artificial Analysis Intelligence IndexSource 40.0%54.8%GPT-5.5 leads
AA-GPQA DiamondSource 87.5%93.5%GPT-5.5 leads
AA-HLESource 26.6%44.3%GPT-5.5 leads
AA-Omniscience IndexSource -18.7%20.1%GPT-5.5 leads
AA-Omniscience AccuracySource 37.5%56.9%GPT-5.5 leads
AA-Omniscience Hallucination RateSource 89.8%85.5%GPT-5.5 leads
GPQA-DSource 93.6%Not comparable
MathGPT-5.5 wins
BenchmarkGPT-5.4 miniGPT-5.5Result
FrontierMath v2 (Tiers 1-3)Source 28.280%51.700%GPT-5.5 leads
FrontierMath v2 (Tier 4)Source 2.080%35.400%GPT-5.5 leads
FrontierMath (legacy)Source 51.7%Not comparable
MultimodalGPT-5.4 mini wins
BenchmarkGPT-5.4 miniGPT-5.5Result
MMMU-ProSource 76.6%81.2%GPT-5.5 leads
MMMU-Pro w/ PythonSource 78%83.2%GPT-5.5 leads
AA-MMMU-ProSource 73.3%79.9%GPT-5.5 leads
OfficeQA ProSource 54.1%Not comparable
Design Arena WebsiteSource 1282Not comparable
Inst. Following
BenchmarkGPT-5.4 miniGPT-5.5Result
AA-IFBenchSource 73.3%75.9%GPT-5.5 leads
Frequently Asked Questions (5)

Which is better, GPT-5.4 mini or GPT-5.5?

GPT-5.5 is ahead on BenchLM's BenchAlign leaderboard, 73.51 to 56.77. The biggest single separator in this matchup is FrontierMath v2 (Tier 4), where the scores are 2.080% and 35.400%.

Which is better for knowledge tasks, GPT-5.4 mini or GPT-5.5?

GPT-5.5 has the edge for knowledge tasks in this comparison, averaging 57.8 versus 47.8. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.

Which is better for math, GPT-5.4 mini or GPT-5.5?

GPT-5.5 has the edge for math in this comparison, averaging 47.6 versus 21.7. Inside this category, FrontierMath v2 (Tier 4) is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, GPT-5.4 mini or GPT-5.5?

GPT-5.5 has the edge for agentic tasks in this comparison, averaging 81.6 versus 65.7. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, GPT-5.4 mini or GPT-5.5?

GPT-5.4 mini has the edge for multimodal and grounded tasks in this comparison, averaging 76.6 versus 70.4. Inside this category, AA-MMMU-Pro is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 22, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.