priorsBuilder

Other · released 2026-05-29

Step 3.7 Flash

Measured on 21 benchmarks.

General frontier

20.5

#107 of 136 models · best at high

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

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #201high of 58721.4%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #300high of 526-37.3

Leader: Opus 5.5 · 46.4

All 21 benchmarks

show 19 more

agentic

APEX-Agents-AA

33 results
  1. #23high of 3314.8%

Leader: Gemini 3.5 Flash · 47.1%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #43high of 5029.1%

Leader: Fable 5 · 51.1%

agentic

ITBench-AA

45 results
  1. #34high of 4530.3%

Leader: GPT-5.6 Sol · 56.2%

agentic

τ²-Bench Telecom

393 results
  1. #7high of 39398.5%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #135high of 20612.0%

Leader: Qwen 3.8 Max · 51.3%

coding

SciCode

226 results
  1. #148high of 22643.9%

Leader: Opus 5.5 · 66.9%

coding

Terminal-Bench Hard-AA

387 results
  1. #80high of 38735.6%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #141high of 23839.3%

Leader: Fable 5.1 · 91.4%

composite

AA Intelligence Index v4.3.2

610 results
  1. #233high of 61019.5

Leader: Opus 5.5 · 57.6

economics

GDPval-AA v2.1

288 results
  1. #179high of 288856

Leader: Opus 5.5 · 1866

instruction

IFBench

399 results
  1. #87high of 39967.3%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #215high of 52625.8%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #330high of 52615.0%

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

knowledge

GPQA Diamond

539 results
  1. #185high of 53980.9%

Leader: GPT-6 Astra · 96.3%

knowledge

MMMU-Pro

265 results
  1. #98high of 26575.3%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #175high of 52373.7%

Leader: Kimi K3 · 88.7%

reasoning

CritPt

529 results
  1. #192high of 5292.3%

Leader: Muse Wonder Full (Max) · 32.9%

speed

Latency (Time to First Token)

351 results
  1. #188high of 3512.6s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #68high of 351171.2

Leader: Celeris-1 · 1522.5