VerdictPal
Join

Models / OpenAI

GPT-5.4 Nano

GPT-5.4 · Released 2026-03-17

OpenAI's March 2026 budget tier. $0.20/$1.25 per 1M tokens. Optimized for speed-critical, high-volume tasks.

Cheapest GPT-5.4 for routing, extraction, classification, and high-volume backend tasks.

#17 of 21 on Vals Index · current snapshot
46 Vals Index · 11 benchmark rows · 2 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite52
Context128K
Input / 1M$0.20
Output / 1M$1.3
Knowledge cutoff2026-01-01
Output speed163 t/s
TTFT (default API)3.46s
Statussolid
ReasoningCodingAgenticLong context
Reasoning
82
Coding
47
Agentic
61
Long context
66

64average across 4 tested families

Benchmark placements

Where GPT-5.4 Nano 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
AA output speedPerformance#14/ 69Artificial Analysis2026-07-21
IFBenchReasoning#14/ 5776Artificial Analysis2026-07-21
Vals IndexAgentic#16/ 2046Vals AI — Vals Index2026-06-04
AA time to first tokenPerformance#27/ 69Artificial Analysis2026-07-21
AA-LCRLong context#35/ 6466Artificial Analysis2026-07-21
Artificial Analysis Intelligence IndexReasoning#36/ 6738Artificial Analysis2026-07-21
SciCodeCoding#38/ 6547Artificial Analysis2026-07-21
Terminal-BenchAgentic#38/ 6961Artificial Analysis2026-07-21
τ³-Bench BankingAgentic#39/ 6421Artificial Analysis2026-07-21
Humanity's Last ExamReasoning#43/ 6527Artificial Analysis2026-07-21
GPQA DiamondReasoning#45/ 6982Artificial Analysis2026-07-21
Family context

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

ReasoningGPT-5.4 Nano · 82
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
CodingGPT-5.4 Nano · 47
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGPT-5.4 Nano · 61
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGPT-5.4 Nano · 66
  1. 1
    Kimi K3
    75
  2. 2
    GPT-5.1
    75
  3. 3
    GPT-5.6 Sol
    74
Newest receipts

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

11 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context66%Artificial Analysis2
AA output speedPerformance163 t/sArtificial Analysis2
AA time to first tokenPerformance3.46sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning38.2%Artificial Analysis2
GPQA DiamondReasoning81.7%Artificial Analysis2
Humanity's Last ExamReasoning26.5%Artificial Analysis2
IFBenchReasoning75.92%Artificial Analysis2
SciCodeCoding46.9%Artificial Analysis2
τ³-Bench BankingAgentic21.03%Artificial Analysis2
Terminal-BenchAgentic60.67%Artificial Analysis2
Vals IndexAgentic46.46%Vals AI — Vals Index49
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.

Reasoning82
Coding47
Agentic61
Long context66
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMiniMaxxAIMetaZhipuXiaomiMeituanCohere

Full benchmark ledger

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

11 sourced rows

11 of 32 catalog benchmarks have a sourced row for GPT-5.4 Nano.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. Released GPT-5.4 Nano — budget tier, $0.20/$1.25.
Sources