VerdictPal
Join

Models / OpenAI

GPT 5.4

GPT 5 · Released 2026-03-05

OpenAI's March 2026 model. AA Intelligence Index 56.8 (#5 globally). SWE-bench Verified 80.0%, Terminal-Bench 2.0 81.8% (ForgeCode). ARC-AGI-2 83.3% (Pro variant). 256K context, $7 / $28 per 1M tokens — 44% cheaper than GPT 5.5 on input.

Default for production OpenAI workloads where GPT 5.5 is overkill but GPT 5.2 is too old. The Pro variant leads ARC-AGI-2 among non-OpenAI competitors at 83.3%.

#14 of 168 on AA Index · current snapshot
53 AA Index · 19 benchmark rows · 10 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite62
Context256K
Input / 1M$7.0
Output / 1M$28
Knowledge cutoff2026-01-31
Statusflagship
ChatGPTCursorPerplexityScira AI

Family profile

Best published score in each covered benchmark family.

79/100 avg
Knowledge
90
Reasoning
92
Math
48
Coding
80
Agentic
82
Long context
78
Tool use
84
Human preference
81

8 tested benchmark families

Editor's note
GPT 5.4 is the OpenAI model we recommend when GPT 5.5 is too expensive but the task still needs frontier coding. The ForgeCode variant of Terminal-Bench 2.0 (81.8%) is the most reproducible number we have on the model.
Benchmark placements

Where GPT 5.4 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
FrontierMathMath#4/ 1548Epoch AI FrontierMath Tiers 1-3 (v2) CSV2026-08-15
BFCLTool use#5/ 2084Berkeley Function Calling Leaderboard V42026-04-15
DeepSWECoding#6/ 852DeepSWE leaderboard v1.12026-06-20
LiveBenchReasoning#6/ 2166LiveBench leaderboard2026-04-15
LongBench v2Long context#6/ 1954LongBench v2 leaderboard2026-04-01
SciCodeCoding#6/ 18057Artificial Analysis2026-09-01
MMLUKnowledge#7/ 2290OpenAI GPT-5.4 release2026-03-05
LMArena (Chatbot Arena)Human preference#8/ 1381LMArena leaderboard2026-05-25
SWE-bench VerifiedCoding#12/ 2380SWE-bench official leaderboard2026-05-30
ARC-AGIReasoning#14/ 140ARC-AGI-3 paper (arXiv 2603.24621)2026-03-20
Humanity's Last ExamReasoning#15/ 18044Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#18/ 10640Artificial Analysis2026-09-01
Terminal-BenchAgentic#18/ 18482Terminal-Bench 2.0 leaderboard2026-05-20
AA-LCRLong context#21/ 17978Artificial Analysis2026-09-01
GPQA DiamondReasoning#23/ 18592Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#24/ 18753Artificial Analysis2026-09-01
IFBenchReasoning#32/ 12774Artificial Analysis2026-09-01
AA time to first tokenPerformance#123/ 186Artificial Analysis2026-09-01
AA output speedPerformance#124/ 187Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where GPT 5.4 shows up, it's highlighted.

KnowledgeGPT 5.4 · 90
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningGPT 5.4 · 92
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
MathGPT 5.4 · 48
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    GPT-5 (high)
    99
CodingGPT 5.4 · 80
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGPT 5.4 · 82
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGPT 5.4 · 78
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Tool useGPT 5.4 · 84
  1. 1
    Claude Sonnet 4.6
    89
  2. 2
    GPT 5.5
    87
  3. 3
    Claude Opus 4.8
    87
Human preferenceGPT 5.4 · 81
  1. 1
    GPT 5.5
    95
  2. 2
    Claude Opus 4.8
    92
  3. 3
    Gemini 3.5 Flash
    91
Newest receipts

Benchmark rows added to the public ledger for GPT 5.4 in the last 120 days. Older rows live in the full table below.

14 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context77.67%Artificial Analysis5
AA output speedPerformance0 t/sArtificial Analysis5
AA time to first tokenPerformance0.00sArtificial Analysis5
Artificial Analysis Intelligence IndexReasoning53.1%Artificial Analysis5
GPQA DiamondReasoning92%Artificial Analysis5
Humanity's Last ExamReasoning43.7%Artificial Analysis5
IFBenchReasoning73.95%Artificial Analysis5
SciCodeCoding56.6%Artificial Analysis5
τ³-Bench BankingAgentic39.59%Artificial Analysis5
FrontierMathMath47.6%Epoch AI FrontierMath Tiers 1-3 (v2) CSV22
DeepSWECoding52%DeepSWE leaderboard v1.178
SWE-bench VerifiedCoding80%SWE-bench official leaderboard99
LMArena (Chatbot Arena)Human preference1405LMArena leaderboard104
Terminal-BenchAgentic81.8%Terminal-Bench 2.0 leaderboard109
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge90
Reasoning92
Math48
Coding80
Agentic82
Long context78
Tool use84
Human preference81
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for GPT 5.4. Scores link to the original source; gaps mean no public row exists yet.

19 sourced rows

19 of 37 catalog benchmarks have a sourced row for GPT 5.4.

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Tool use

Performance

Human preference

Changelog
  1. Released GPT 5.4 — AA Index 56.8, Terminal-Bench 2.0 81.8% (ForgeCode).
  2. GPT 5.2 retired for general availability.
Sources