priorsBuilder

OpenAI · released 2026-03-17

GPT-5.4 Mini

Measured on 76 benchmarks.

General frontier

Not ranked in this mix — too little measured data for its benchmarks.

Your mix’s benchmarks

4 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #163xhigh of 58728.1%
  2. #220medium of 58718.6%
  3. #396none of 5875.9%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #233xhigh of 526-18.9
  2. #239medium of 526-20.6
  3. #340none of 526-45.4

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #115high of 19632.0%
  2. #116none of 19632.0%
  3. #119xhigh of 19631.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #59xhigh of 709.8%

Leader: GPT-6.1 Sol · 100.0%

All 76 benchmarks

show 72 more

agentic

AA-AnalystAgent

44 results
  1. #35xhigh of 4415.0%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #159xhigh of 224719

Leader: Sonnet 5.5 · 1823

agentic

APEX-Agents-AA

33 results
  1. #14xhigh of 3328.2%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench-AA

219 results
  1. #119xhigh of 21926.2%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #36xhigh of 5034.6%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #75xhigh of 790.0%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #27xhigh of 4535.2%
  2. #40none of 4518.6%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #67xhigh of 7812.5%

Leader: Muse Spark 1.3 · 55.3%

agentic

Tax Agent Bench

69 results
  1. #57xhigh of 698.8%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #42xhigh of 472.5%

Leader: Opus 5.5 · 65.2%

agentic

τ²-Bench Telecom

393 results
  1. #111xhigh of 39383.3%
  2. #231medium of 39336.5%
  3. #297none of 39323.4%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #81xhigh of 20625.6%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #85high of 1221397

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #60xhigh of 7712.9%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #17xhigh of 2924.3%

Leader: GPT-5.5 · 70.0%

coding

FrontierCode 1.1 Extended

132 results
  1. #108xhigh of 13243.0%
  2. #118high of 13237.2%
  3. #119medium of 13235.1%
  4. #130low of 13220.6%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

132 results
  1. #107xhigh of 13227.0%
  2. #117high of 13223.1%
  3. #120medium of 13220.8%
  4. #131low of 1329.3%

Leader: Opus 5.5 · 54.6%

coding

LiveCodeBench (Vals)

127 results
  1. #62xhigh of 12781.5%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #38xhigh of 623.0%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #55xhigh of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #40xhigh of 6254.4%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #84xhigh of 22652.1%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #50xhigh of 8573.0%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #26high of 6344.9%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #49high of 7454.7%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #20xhigh of 38752.3%
  2. #90medium of 38734.1%
  3. #177none of 38718.2%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #101xhigh of 23859.2%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #116xhigh of 2162.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #53xhigh of 10548.0%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #46xhigh of 12760.3%
  2. #97none of 12737.9%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #176xhigh of 61024.1
  2. #231medium of 61019.7
  3. #350none of 61011.1

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #71high of 2201447

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #59high of 2201466

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #44xhigh of 4733.2%

Leader: Gemini 4 · 68.9%

composite

Vals Multimodal Index v1.2

33 results
  1. #21xhigh of 3354.2%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #83xhigh of 177$0.45

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #118xhigh of 177$1,524

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #41xhigh of 721436

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #63xhigh of 11760.9%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #56xhigh of 7445.4%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #53xhigh of 7845.4%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #102xhigh of 21013.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #157xhigh of 2881012
  2. #205none of 288589

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #49xhigh of 8463.5%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #75xhigh of 12571.2%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #42xhigh of 39973.3%
  2. #104medium of 39964.8%
  3. #266none of 39938.8%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #138xhigh of 52637.5%
  2. #146medium of 52636.5%
  3. #218none of 52625.7%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #399medium of 52610.2%
  2. #405xhigh of 5269.8%
  3. #505none of 5264.4%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

CaseLaw (Vals)

53 results
  1. #47xhigh of 5351.7%

Leader: Grok 4.3 · 79.3%

knowledge

Global-MMLU-Lite

118 results
  1. #19xhigh of 11889.4%

Leader: Gemini 3.1 Pro · 93.2%

knowledge

GPQA Diamond

539 results
  1. #97xhigh of 53987.5%
  2. #171medium of 53982.3%
  3. #355none of 53960.6%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #59xhigh of 12283.1%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MMLU Pro (Vals)

122 results
  1. #61xhigh of 12284.6%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #47xhigh of 8179.2%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #121xhigh of 26573.3%
  2. #135medium of 26571.2%
  3. #198none of 26560.5%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #135xhigh of 52377.0%
  2. #236medium of 52367.0%
  3. #355none of 52337.0%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #80xhigh of 943.9%

Leader: Sonnet 5.5 · 75.0%

math

AIME (Vals)

82 results
  1. #10xhigh of 8295.6%

Leader: Gemini 3.1 Pro · 98.1%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #48xhigh of 11551.2%
  2. #78low of 11524.6%
  3. #97none of 11517.2%

Leader: GPT-6.1 Sol · 93.7%

other

Arena Vision

116 results
  1. #51high of 1161251

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #26xhigh of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #135xhigh of 21563.7%
  2. #141high of 21558.0%
  3. #163medium of 21540.8%
  4. #192low of 21513.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #124xhigh of 21618.9%
  2. #134high of 21613.2%
  3. #160medium of 2164.4%
  4. #190low of 2161.1%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #116xhigh of 52910.0%
  2. #179medium of 5292.9%
  3. #368none of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Roboflow Visual Reasoning

102 results
  1. #54high of 10264.0%
  2. #71low of 10257.0%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #15xhigh of 8050.8%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

172 results
  1. #26xhigh of 1720.3%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #69xhigh of 18299.8%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #18none of 3510.7s
  2. #268medium of 35116s
  3. #335xhigh of 351128s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #24xhigh of 351233.1
  2. #49medium of 351193.9
  3. #61none of 351182.6

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #61xhigh of 1781.0s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #9xhigh of 176126.0

Leader: Muse Spark 1.1 · 204.5