priorsBuilder

OpenAI · released 2025-08-07

GPT-5 Mini

Measured on 63 benchmarks.

General frontier

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

Your mix’s benchmarks

3 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #200high of 58721.5%
  2. #237medium of 58715.9%
  3. #437minimal of 5875.1%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #199medium of 526-11.0
  2. #225high of 526-17.3
  3. #394minimal of 526-53.0

Leader: Opus 5.5 · 46.4

math

FrontierMath Tier 4 (v2)

70 results
  1. #57high of 7012.2%

Leader: GPT-6.1 Sol · 100.0%

All 63 benchmarks

show 60 more

agentic

AA-Briefcase v1.1

224 results
  1. #181high of 224428

Leader: Sonnet 5.5 · 1823

agentic

AutomationBench-AA

219 results
  1. #154high of 2196.5%

Leader: Gemini 4 · 77.5%

agentic

Terminal-Bench 2.0

40 results
  1. #28high of 4034.8%

Leader: GPT-5.5 · 84.7%

agentic

Vending-Bench 2

67 results
  1. #67high of 67$-31.18

Leader: GPT-6 Astra · $15,515

agentic

τ²-Bench Telecom

393 results
  1. #155medium of 39371.1%
  2. #163high of 39368.4%
  3. #248minimal of 39331.9%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #117high of 20615.5%

Leader: Qwen 3.8 Max · 51.3%

coding

LiveCodeBench

282 results
  1. #17high of 28283.8%
  2. #69medium of 28269.2%
  3. #116minimal of 28254.5%

Leader: Gemini 3 Pro · 91.7%

coding

LiveCodeBench (Vals)

127 results
  1. #21high of 12786.6%

Leader: Fable 5.1 · 90.5%

coding

SciCode

226 results
  1. #176high of 22639.0%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #74high of 8560.8%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #49high of 6327.0%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench Hard-AA

387 results
  1. #94high of 38733.3%
  2. #120medium of 38728.8%
  3. #198minimal of 38714.4%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #215high of 2383.7%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #169high of 2160.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #83high of 10514.2%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #56high of 12752.7%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #220medium of 61020.6
  2. #263high of 61016.8
  3. #376minimal of 6109.9

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #130high of 2201390

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #140high of 2201372

Leader: Gemini 4 · 1553

cost

AA Intelligence Index Cost per Task

177 results
  1. #24high of 177$0.05

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #41high of 177$189

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #59high of 721123

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #69high of 11760.2%

Leader: Opus 5 · 73.2%

economics

GDP.pdf

210 results
  1. #141high of 2108.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #188high of 288766

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #29high of 8466.9%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #22high of 12575.2%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #32high of 39975.4%
  2. #59medium of 39971.2%
  3. #202minimal of 39945.6%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #223high of 52625.0%
  2. #254medium of 52623.1%
  3. #304minimal of 52619.2%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #89medium of 52655.7%
  2. #142high of 52643.6%
  3. #390minimal of 52610.7%

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

knowledge

CaseLaw (Vals)

53 results
  1. #4high of 5368.5%

Leader: Grok 4.3 · 79.3%

knowledge

Global-MMLU-Lite

118 results
  1. #27high of 11887.4%

Leader: Gemini 3.1 Pro · 93.2%

knowledge

GPQA Diamond

539 results
  1. #166high of 53982.8%
  2. #190medium of 53980.3%
  3. #300minimal of 53968.7%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #68high of 12280.3%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #45high of 9443.0%

Leader: Opus 5 · 63.6%

knowledge

MedQA (Vals)

80 results
  1. #7high of 8096.1%

Leader: o1 · 96.5%

knowledge

MedScribe

96 results
  1. #51high of 9680.6%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #74high of 12282.2%

Leader: Fable 5.1 · 92.4%

knowledge

MMLU-Pro

285 results
  1. #40high of 28583.7%
  2. #54medium of 28582.8%
  3. #125minimal of 28577.5%

Leader: Gemini 3 Pro · 89.8%

knowledge

MMMU Pro (Vals)

81 results
  1. #48high of 8178.9%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #145high of 26570.1%
  2. #156medium of 26568.8%
  3. #206minimal of 26558.4%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #187medium of 52372.3%
  2. #188high of 52372.3%
  3. #348minimal of 52339.3%

Leader: Kimi K3 · 88.7%

math

AIME (Vals)

82 results
  1. #24high of 8291.5%

Leader: Gemini 3.1 Pro · 98.1%

math

AIME 2025

228 results
  1. #22high of 22890.7%
  2. #43medium of 22885.0%
  3. #122minimal of 22846.7%

Leader: GPT-5.2 · 99.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #50high of 11546.7%
  2. #94low of 11518.2%
  3. #107minimal of 1156.0%

Leader: GPT-6.1 Sol · 93.7%

math

MATH 500 (Vals)

48 results
  1. #6high of 4894.8%

Leader: Gemini 3 Pro · 96.4%

math

MGSM (Vals)

62 results
  1. #14high of 6292.6%

Leader: Opus 4.5 · 95.2%

other

Arena Vision

116 results
  1. #88high of 1161183

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #31high of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #148high of 21554.3%
  2. #169medium of 21537.3%
  3. #180low of 21526.3%
  4. #206minimal of 2155.3%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #159high of 2164.4%
  2. #161medium of 2164.0%
  3. #180minimal of 2161.7%
  4. #196low of 2160.8%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #207medium of 5291.4%
  2. #324high of 5290.0%
  3. #386minimal of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #76high of 13581.8%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #51high of 8043.0%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #27high of 1710.3%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #31high of 181100.0%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #28minimal of 3510.8s
  2. #252medium of 35111s
  3. #310high of 35145s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #74high of 351164.3
  2. #76minimal of 351162.6
  3. #83medium of 351157.1

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #153high of 1773.5s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #31high of 176100.3

Leader: Muse Spark 1.1 · 204.3