← All models

GPT-5.4 mini (xhigh)

deprecated by benchmark source
OpenAI · released 2026-03-17 · 6 offers

Context 400K tokens

Top 3 cheapest providers (Adjusted $/task)

The same list price can give a different adjusted $/task (caching, token efficiency) — click a price for its inputs.

Within the active global provider, residency and confidentiality filters.

#ProviderAdjusted $/task
1Azure
2Azure AI Foundry
3T-Systems LLM Hub

Composite

4 of 7 inputs · 2 from the model family6 radar axes: DesignArena's two boards share one

61.6

AA CodingCoding Agent v1.4AA IntelligenceAA AgenticEpoch ECISoftware ECIDesignArenaAA Coding: percentile 63AA Intelligence: percentile 79Epoch ECI: percentile 49Software ECI: percentile 46

AA Coding 56.1Coding Agent v1.4 AA Intelligence 24.6AA Agentic Epoch ECI 149.0Software ECI 151.3DesignArena —/—

Radar: percentile among all models measured on each input; a gap means not measured.

Benchmark sheet

27 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.

Composite attachments (used in the score, not counted as exact benchmarks):
  • Epoch ECI · attached
  • Software ECI · attached
Compare this model →

Agentic

  • AA-AnalystAgent2715.0%

    AA-AnalystAgent published 2026-09-10 · Published board — Tests analyst tasks using agentic Python execution across fourteen domains.

  • APEX-Agents-AA6228.2%

    APEX-Agents-AA published 2026-09-10 · Published board — Tests professional-service tasks that require agents to produce locally graded deliverables.

  • AA-Briefcase33719 Elo

    AA-Briefcase published 2026-09-10 · Published board — Tests multi-week professional knowledge-work projects with linked tasks and large source collections.

  • GDPval-AA v2561,095 Elo

    GDPval-AA v2 v2 · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.

  • Harvey LAB-AA1260.7%

    Harvey LAB-AA published 2026-09-10 · Published board — Tests legal-work deliverables across practice areas on Harvey's private task set. Artificial Analysis' run, graded by one LLM judge against task rubrics — not the same run or scale as Vals AI's HLAB row.

  • ITBench-AA5335.2%

    ITBench-AA published 2026-09-10 · Published board — Tests root-cause diagnosis from offline Kubernetes incident snapshots.

  • Terminal-Bench Hard (AA)9652.3%

    Terminal-Bench Hard (AA) pinned revision 74221fb · Published board — Tests a pinned 44-task hard subset of terminal-based work using Terminus 2.

  • Terminal-Bench v2.1 (AA)5759.2%

    Terminal-Bench v2.1 (AA) v2.1 · Published board — Tests terminal-based work on the 89-task verified refresh using Terminus 2.

  • Terminal-Bench v4.0 (AA)532.0%

    Terminal-Bench v4.0 (AA) v4.0 · Published board — Tests terminal-based work on the 66-task release using mini-SWE-agent v2.4.6.

Coding

  • SciCode (AA subproblems) v1.0.16852.1%

    SciCode (AA subproblems) v1.0.1 · Published board — Tests scientific Python programming with scientist-annotated background information.

  • FrontierCode 1.1 Main (Cognition) v1.1no percentile27.0%

    FrontierCode 1.1 Main (Cognition) v1.1 · codex — Whether a maintainer would merge the agent's pull request, on tasks crafted by open-source maintainers and graded with tests, rubrics and verifiers.

  • AA Coding Index6356.1

    AA Coding Index published 2026-09-19 · Published board — Artificial Analysis publishes this board without a version number; we keep the date each result was retained.

Efficiency

Instruction-following

Knowledge

Long-context

  • GDP.pdf (AA)6113.4%

    GDP.pdf (AA) published 2026-09-10 · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.

  • AA-LCR v1.18177.0%

    AA-LCR v1.1 v1.1 · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.

  • MLCR-AA163.9%

    MLCR-AA published 2026-09-10 · Published board — Tests medical-record synthesis and reasoning across long, fragmented case documents.

Reasoning

Safety/Alignment

Science

Tool-use

  • AutomationBench-AA v1.0.65226.2%

    AutomationBench-AA v1.0.6 · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.

  • EnterpriseOps-Gym-AA3034.6%

    EnterpriseOps-Gym-AA published 2026-09-10 · Published board — Tests enterprise workflows through MCP tools against resettable application environments.

  • τ²-Bench Telecom (AA)7483.3%

    τ²-Bench Telecom (AA) published 2026-09-10 · Published board — Tests dual-control telecom agents that coordinate tool use with a simulated user.

  • τ³-Banking (AA) v1.0.16225.6%

    τ³-Banking (AA) v1.0.1 · Published board — Tests banking support agents that retrieve policies and change account state through tools.

Vision

Unusual results

1 threshold-crossing signal flagged · 26 eligible benchmark families

How flags are calculated

Heuristic screen, not statistical significance: benchmark families are correlated and source uncertainty is unknown. Peer evidence needs ≥ 20 independently measured matched configurations from ≥ 10 distinct model families. A flag needs a directed population z-score of magnitude ≥ 1.5 and a gap of ≥ 1.5 from the leave-one-benchmark-family-out mean z in the same direction, over ≥ 5 other benchmark families.

  • unusually strong Terminal-Bench Hard (AA)
    Why
    Observed: 0.52273 fraction
    Peer mean 0.18472 · peer sd 0.16951 · n 432 · families 315
    Directed z 1.994 · baseline z 0.243 · gap 1.751 · profile n 25
    0.52273 fractionmeasuredobserved 2026-09-10artificialanalysis.ai
    Evidence
    Axis: Terminal-Bench Hard (AA) · pinned revision 74221fb · Published board
    Exact value: 0.522727272727273 fraction
    Observed: 2026-09-10T21:47:16.627Z · publication date: not recorded
    Observation id: aa:ba242e40-83b7-4cd3-a0e0-b56237984914:terminalbenchHard
Missing coverage · 116 benchmark versions

No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.

Variants / reasoning settings

Artificial Analysis snapshot 2026-09-19 · Data: Artificial Analysis · Data: Epoch AI (CC BY)

VariantAA CodingAA Intelligence
GPT-5.4 mini (xhigh)56.124.6
GPT-5.4 mini (medium)19.7
GPT-5.4 mini (Non-Reasoning)11.1

SUBSCRIPTION PLAN

GitHub Copilot

Current usage-based billing · model token cost is converted to AI Credits at 1 credit = $0.01.

Input $0.750 / 1M
Cached input $0.075 / 1M
Output $4.50 / 1M
Status GA

Legacy annual Pro/Pro+ request billing only.

Multiplier 6.00×
Effective cost $0.240 / request

GitHub marks this multiplier as subject to change.

Token offers by platform · 4 offers (Adjusted $/task)

Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.

4 offers within the active global filters; “—” means the catalog is active but no public token price is available.

OpenRouter (2)

Azure
Azure

Azure AI Foundry (1)

Azure AI Foundry

T-Systems LLM Hub (1)

T-Systems LLM Hub
GPT-5.4 mini (xhigh) — benchmarks & cost | Benchmark Heaven