ARC-AGIVerified
83.3%
ARC-AGI-2 semi-private
Models / OpenAI
GPT-5.4 Pro
GPT-5.4 · Released 2026-03-05
OpenAI's March 2026 premium tier. $30/$180 per 1M tokens. Highest-precision variant for expensive-but-important workloads.
Highest-precision GPT-5.4 for financial analysis, scientific research, and high-stakes decision support.
Composite55
Context128K
Input / 1M$30
Output / 1M$180
Knowledge cutoff2026-01-01
Statussolid
Not enough families covered yet.Add more benchmark rows for GPT-5.4 Pro to see the family profile.
Benchmark placements
Per-benchmark positions
Where GPT-5.4 Pro 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.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| ARC-AGI | Reasoning | #3/ 14 | 83 | ARC Prize leaderboard | 2026-05-15 |
| FrontierMath | Math | #3/ 15 | 50 | Epoch AI FrontierMath Tiers 1-3 (v2) CSV | 2026-08-15 |
| CritPt | Reasoning | #4/ 12 | 30 | Artificial Analysis — CritPt | 2026-09-01 |
| AA time to first token | Performance | #137/ 186 | Artificial Analysis | 2026-09-01 | |
| AA output speed | Performance | #138/ 187 | Artificial Analysis | 2026-09-01 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where GPT-5.4 Pro shows up, it's highlighted.
Knowledge
- 191
- 291
- 391
ReasoningGPT-5.4 Pro · 83
- 195
- 295
- 395
MathGPT-5.4 Pro · 50
- 199
- 299
- 399
Coding
- 196
- 296
- 396
Agentic
- 1100
- 2100
- 3100
Long context
- 183
- 283
- 381
Tool use
- 189
- 287
- 387
Safety
- 192
- 292
- 391
Human preference
- 195
- 292
- 391
Newest receipts
5 recentLast 120 days
Benchmark rows added to the public ledger for GPT-5.4 Pro in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| CritPt | Reasoning | 30% | Artificial Analysis — CritPt | 5 |
| AA output speed | Performance | 0 t/s | Artificial Analysis | 5 |
| AA time to first token | Performance | 0.00s | Artificial Analysis | 5 |
| FrontierMath | Math | 50% | Epoch AI FrontierMath Tiers 1-3 (v2) CSV | 22 |
| ARC-AGI | Reasoning | 83.3% | ARC Prize leaderboard | 114 |
Family coverage
Best score per family
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.
Price vs. performance
Where GPT-5.4 Pro sits on the scatter
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 Pro. Scores link to the original source; gaps mean no public row exists yet.
5 of 37 catalog benchmarks have a sourced row for GPT-5.4 Pro.
Reasoning
CritPtVerified
30%
CritPt composite (AA run)
Math
FrontierMathVerified
50%
FrontierMath Tiers 1-3 (v2) best Epoch run
Performance
AA output speedVerified
0 t/s
Median output tokens/s (1k prompt, default provider)
AA time to first tokenVerified
0.00s
Median time to first token (1k prompt, default provider)
Changelog
Version history
- Released GPT-5.4 Pro — premium tier, $30/$180.
Sources
Where the numbers come from
- OpenAI announcement· checked 2026-06-02