priorsBuilder

Google · released 2026-06-03

Gemma 4 12B

Measured on 16 benchmarks across 2 reasoning efforts.

General frontier

15.7

#123 of 136 models

includes estimates: 24% of the mix is measured for this model

Reasoning efforts in your mix

default15.7#293 of 313
none12.1#303 of 313

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #243default of 58715.7%
  2. #381none of 5876.3%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #392default of 526-52.7
  2. #393none of 526-52.8

Leader: Opus 5.5 · 46.4

All 16 benchmarks

show 14 more

agentic

τ²-Bench Telecom

393 results
  1. #235default of 39336.3%
  2. #249none of 39331.9%

Leader: GLM 5.2 · 99.1%

coding

Terminal-Bench Hard-AA

387 results
  1. #172default of 38718.2%
  2. #214none of 38711.4%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #163default of 23827.3%

Leader: Fable 5.1 · 91.4%

composite

AA Intelligence Index v4.3.2

610 results
  1. #294default of 61014.2
  2. #387none of 6109.4

Leader: Opus 5.5 · 57.6

economics

GDPval-AA v2.1

288 results
  1. #225default of 288427

Leader: Opus 5.5 · 1866

instruction

IFBench

399 results
  1. #39default of 39973.5%
  2. #206none of 39945.2%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #386default of 52615.6%
  2. #448none of 52611.9%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #227none of 52626.6%
  2. #283default of 52619.0%

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

knowledge

GPQA Diamond

539 results
  1. #244default of 53975.3%
  2. #324none of 53966.1%

Leader: GPT-6 Astra · 96.3%

knowledge

MMMU-Pro

265 results
  1. #148default of 26569.7%
  2. #192none of 26562.0%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #256default of 52363.7%
  2. #361none of 52335.0%

Leader: Kimi K3 · 88.7%

reasoning

CritPt

529 results
  1. #334default of 5290.0%
  2. #391none of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

speed

Latency (Time to First Token)

351 results
  1. #183default of 3512.5s
  2. #184none of 3512.5s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #131none of 351113.6
  2. #139default of 351111.6

Leader: Celeris-1 · 1522.5