priorsBuilder

OpenAI · released 2025-08-07

GPT-5

Measured on 63 benchmarks.

General frontier

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

Your mix’s benchmarks

5 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #159high of 58728.5%
  2. #178medium of 58725.4%
  3. #212low of 58719.6%
  4. #370none of 5876.6%
  5. #392minimal of 5876.0%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #176high of 526-8.7
  2. #194low of 526-10.8
  3. #196medium of 526-10.9
  4. #285minimal of 526-33.8

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #139medium of 19621.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #48high of 7022.0%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #44high of 9756.7%

Leader: Opus 5.5 · 88.4%

All 63 benchmarks

show 58 more

agentic

Butter-Bench

5 results
  1. #3high of 530.0%

Leader: Gemini 2.5 Pro · 40.0%

agentic

Terminal-Bench 2.0

40 results
  1. #16high of 4049.6%

Leader: GPT-5.5 · 84.7%

agentic

τ²-Bench Telecom

393 results
  1. #86medium of 39386.5%
  2. #101high of 39384.8%
  3. #104low of 39384.2%
  4. #169minimal of 39367.0%
  5. #373none of 3930.0%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #90high of 20622.1%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #76medium of 1221418

Leader: Opus 5.5 · 1813

coding

LiveCodeBench

282 results
  1. #15high of 28284.6%
  2. #40low of 28276.3%
  3. #62medium of 28270.3%
  4. #113minimal of 28255.8%
  5. #117none of 28254.3%

Leader: Gemini 3 Pro · 91.7%

coding

LiveCodeBench (Vals)

127 results
  1. #28high of 12785.9%

Leader: Fable 5.1 · 90.5%

coding

SWE-bench Verified (Vals)

85 results
  1. #64high of 8569.0%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #38high of 6337.1%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench Hard-AA

387 results
  1. #66medium of 38737.9%
  2. #97high of 38732.6%
  3. #131low of 38726.5%
  4. #176minimal of 38718.2%
  5. #205none of 38712.9%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #149high of 23835.2%

Leader: Fable 5.1 · 91.4%

coding

Vibe Code Bench v1.1

105 results
  1. #74high of 10520.1%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #89high of 12739.8%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #187high of 61023.0
  2. #190medium of 61022.9
  3. #215low of 61020.8
  4. #343minimal of 61011.4
  5. #364none of 61010.4

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #86high of 2201435

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #78high of 2201449

Leader: Gemini 4 · 1553

creative

VoxelBench (image)

16 results
  1. #11high of 161226

Leader: GPT-6.1 Sol · 2204

creative

VoxelBench (text)

55 results
  1. #46high of 551292

Leader: GPT-6 Astra · 2733

economics

CorpFin (Vals)

117 results
  1. #60high of 11761.1%

Leader: Opus 5 · 73.2%

economics

GDPval-AA v2.1

288 results
  1. #174high of 288917

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #39high of 8465.5%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #47high of 12573.4%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #44high of 39973.1%
  2. #65medium of 39970.6%
  3. #92low of 39966.6%
  4. #201minimal of 39945.6%
  5. #207none of 39945.0%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #121high of 52640.3%
  2. #126medium of 52639.5%
  3. #134low of 52638.1%
  4. #181minimal of 52629.6%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #268low of 52621.0%
  2. #302high of 52617.8%
  3. #313medium of 52616.8%
  4. #404minimal of 5269.9%

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

knowledge

Arena Search

28 results
  1. #26high of 281133

Leader: GPT-5.6 Sol · 1257

knowledge

CaseLaw (Vals)

53 results
  1. #6high of 5366.5%

Leader: Grok 4.3 · 79.3%

knowledge

Global-MMLU-Lite

118 results
  1. #7high of 11890.7%
  2. #12medium of 11890.4%
  3. #42minimal of 11884.4%

Leader: Gemini 3.1 Pro · 93.2%

knowledge

GPQA Diamond

539 results
  1. #129high of 53985.4%
  2. #149medium of 53984.2%
  3. #187low of 53980.8%
  4. #303none of 53968.6%
  5. #311minimal of 53967.3%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #46high of 12285.6%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #20high of 9449.6%

Leader: Opus 5 · 63.6%

knowledge

MedQA (Vals)

80 results
  1. #4high of 8096.3%

Leader: o1 · 96.5%

knowledge

MedScribe

96 results
  1. #41high of 9683.7%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #40high of 12286.5%

Leader: Fable 5.1 · 92.4%

knowledge

MMLU-Pro

285 results
  1. #12high of 28587.1%
  2. #14medium of 28586.7%
  3. #21low of 28586.0%
  4. #63none of 28582.0%
  5. #86minimal of 28580.6%

Leader: Gemini 3 Pro · 89.8%

knowledge

MMMU Pro (Vals)

81 results
  1. #39high of 8181.5%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #107medium of 26574.3%
  2. #109high of 26574.2%
  3. #117low of 26573.8%
  4. #190minimal of 26562.1%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #116high of 52378.2%
  2. #147medium of 52376.0%
  3. #245none of 52365.0%

Leader: Kimi K3 · 88.7%

math

AIME (Vals)

82 results
  1. #14high of 8293.4%

Leader: Gemini 3.1 Pro · 98.1%

math

AIME 2025

228 results
  1. #12high of 22894.3%
  2. #17medium of 22891.7%
  3. #50low of 22883.0%
  4. #119none of 22848.3%
  5. #153minimal of 22831.7%

Leader: GPT-5.2 · 99.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #44high of 11555.4%
  2. #60low of 11537.2%
  3. #93minimal of 11518.2%

Leader: GPT-6.1 Sol · 93.7%

math

MATH 500 (Vals)

48 results
  1. #3high of 4896.0%

Leader: Gemini 3 Pro · 96.4%

math

MATH-500

162 results
  1. #1high of 16299.4%
  2. #4medium of 16299.1%
  3. #8low of 16298.7%
  4. #72minimal of 16286.1%

Leader: this model · 99.4%

math

MGSM (Vals)

62 results
  1. #12high of 6292.8%

Leader: Opus 4.5 · 95.2%

other

Arena Vision

116 results
  1. #68high of 1161226

Leader: Fable 5 · 1308

other

Poker Agent

17 results
  1. #2high of 171103

Leader: GPT-5.2 · 1132

other

T3 Code usage share

85 results
  1. #40high of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #130high of 21565.7%
  2. #146medium of 21556.2%
  3. #159low of 21544.0%
  4. #201minimal of 2156.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #139high of 2169.9%
  2. #144medium of 2167.5%
  3. #176low of 2161.9%
  4. #212minimal of 2160.0%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #145high of 5295.7%
  2. #219low of 5291.1%
  3. #311medium of 5290.0%
  4. #363minimal of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #17high of 13586.0%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #46high of 8043.7%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

172 results
  1. #33high of 1720.3%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #33high of 182100.0%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #79minimal of 3511.1s
  2. #247low of 3519.4s
  3. #296medium of 35130s
  4. #322high of 35168s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #149low of 351109.4
  2. #153high of 351106.5
  3. #160medium of 351101.9
  4. #181minimal of 35194.0

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #167high of 1784.5s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #75high of 17672.0

Leader: Muse Spark 1.1 · 204.5