priorsBuilder

xAI · released 2025-02-19

Grok 3 Mini

Measured on 33 benchmarks.

General frontier

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

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #294high of 58711.0%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #166high of 526-6.8

Leader: Opus 5.5 · 46.4

All 33 benchmarks

show 31 more

agentic

τ²-Bench Telecom

393 results
  1. #64high of 39390.4%

Leader: GLM 5.2 · 99.1%

coding

LiveCodeBench

282 results
  1. #64high of 28269.6%

Leader: Gemini 3 Pro · 91.7%

coding

LiveCodeBench (Vals)

127 results
  1. #78high of 12776.2%
  2. #95low of 12766.3%

Leader: Fable 5.1 · 90.5%

coding

Terminal-Bench Hard-AA

387 results
  1. #178high of 38717.4%

Leader: GPT-5.6 Sol · 65.9%

coding

WeirdML v2

127 results
  1. #81high of 12742.6%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #289high of 61014.6

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #144high of 2201364

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #150high of 2201350

Leader: Gemini 4 · 1553

economics

CorpFin (Vals)

117 results
  1. #59high of 11761.1%
  2. #75low of 11759.5%

Leader: Opus 5 · 73.2%

economics

TaxEval (Vals)

125 results
  1. #54low of 12573.0%
  2. #77high of 12571.1%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #197high of 39945.9%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #399high of 52615.1%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #27high of 52674.2%

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

knowledge

GPQA Diamond

539 results
  1. #200high of 53979.1%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #71high of 12279.3%
  2. #85low of 12273.0%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedQA (Vals)

80 results
  1. #43high of 8090.1%
  2. #48low of 8088.7%

Leader: o1 · 96.5%

knowledge

MMLU Pro (Vals)

122 results
  1. #77high of 12281.4%
  2. #85low of 12280.0%

Leader: Fable 5.1 · 92.4%

knowledge

MMLU-Pro

285 results
  1. #55high of 28582.8%

Leader: Gemini 3 Pro · 89.8%

long-context

AA-LCR v1.1

523 results
  1. #287high of 52355.3%

Leader: Kimi K3 · 88.7%

math

AIME (Vals)

82 results
  1. #38high of 8285.0%
  2. #53low of 8270.6%

Leader: Gemini 3.1 Pro · 98.1%

math

AIME 2025

228 results
  1. #45high of 22884.7%

Leader: GPT-5.2 · 99.0%

math

MATH 500 (Vals)

48 results
  1. #11high of 4894.2%
  2. #44low of 4870.2%

Leader: Gemini 3 Pro · 96.4%

math

MATH-500

162 results
  1. #3high of 16299.2%

Leader: GPT-5 · 99.4%

math

MGSM (Vals)

62 results
  1. #33high of 6290.4%
  2. #34low of 6290.4%

Leader: Opus 4.5 · 95.2%

reasoning

ARC-AGI-1

215 results
  1. #189low of 21516.5%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #200low of 2160.4%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #256high of 5290.6%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #57high of 13583.1%
  2. #74low of 13581.9%

Leader: Fable 5 · 88.6%

reasoning

SimpleQA Verified

47 results
  1. #38high of 4720.2%

Leader: Gemini 3.1 Pro · 77.5%

speed

Latency (Time to First Token)

351 results
  1. #128high of 3511.7s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #235high of 35167.6

Leader: Celeris-1 · 1522.5