priorsBuilder

Zhipu · released 2026-06-13

GLM 5.2

Measured on 74 benchmarks across 4 reasoning efforts.

General frontier

44.9

#21 of 32 models · best at xhigh

Reasoning efforts in your mix

xhigh44.9#69 of 129
high36.6#100 of 129
max36.5#101 of 129
none19.1#126 of 129

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #75max of 58741.1%
  2. #318none of 5879.8%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #108max of 5264.4
  2. #164none of 526-6.6

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #124xhigh of 19631.0%
  2. #154none of 19617.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #56max of 7043.8%
  2. #60high of 7036.3%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #37max of 7029.3%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #41max of 9758.8%

Leader: Opus 5.5 · 88.4%

All 74 benchmarks

show 68 more

agentic

AA-Briefcase v1.1

224 results
  1. #84max of 2241229

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #21max of 503.2%

Leader: Opus 5.5 · 15.5%

agentic

APEX-Agents-AA

33 results
  1. #8max of 3333.7%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench (Zapier)

115 results
  1. #82max of 1157.8%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #115max of 21928.4%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #21max of 5042.7%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #28max of 797.1%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #14max of 4542.7%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #43max of 7831.3%

Leader: Muse Spark 1.3 · 55.3%

agentic

PostTrainBench v1.1

12 results
  1. #7max of 1231.7%

Leader: Fable 5 · 41.8%

agentic

SkillsBench

34 results
  1. #26max of 3445.1%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Vending-Bench 2

67 results
  1. #11max of 67$8,314

Leader: GPT-6 Astra · $15,515

agentic

τ²-Bench Telecom

393 results
  1. #1max of 39399.1%

Leader: this model · 99.1%

agentic

τ³-Banking

206 results
  1. #50max of 20634.6%
  2. #110none of 20616.7%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #26max of 1221603

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #33max of 7737.9%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #11max of 2941.5%

Leader: GPT-5.5 · 70.0%

coding

FrontierCode 1.1 Extended

132 results
  1. #111none of 13240.1%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

132 results
  1. #112none of 13224.5%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v1

17 results
  1. #5max of 1767.5%

Leader: Fable 5 · 88.2%

coding

LiveCodeBench (Vals)

127 results
  1. #90max of 12769.5%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #30max of 629.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #29max of 620.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #33max of 6262.6%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #95max of 22651.2%

Leader: Opus 5.5 · 66.9%

coding

SWE Atlas-QnA

22 results
  1. #5max of 2248.1%

Leader: Opus 5 · 63.2%

coding

SWE-bench Verified (Vals)

85 results
  1. #17max of 8582.8%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #32max of 7467.8%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #22max of 38750.8%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #56max of 23877.9%
  2. #116none of 23851.7%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #123max of 2161.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #45max of 10564.0%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #30xhigh of 12770.1%
  2. #33max of 12767.3%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #98max of 61033.7
  2. #197none of 61022.4

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #32max of 2201475

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #45max of 2201475

Leader: Gemini 4 · 1553

cost

AA Intelligence Index Cost per Task

177 results
  1. #134max of 177$1.47

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #138max of 177$2,097

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #32max of 721599

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (text)

55 results
  1. #29max of 551561

Leader: GPT-6 Astra · 2733

economics

CorpFin (Vals)

117 results
  1. #22max of 11766.1%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #32max of 7461.5%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #44max of 7849.7%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #125max of 21010.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #79max of 2881371
  2. #106none of 2881247

Leader: Opus 5.5 · 1866

economics

TaxEval (Vals)

125 results
  1. #48max of 12573.3%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #41max of 39973.3%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #235max of 52624.3%
  2. #293none of 52620.3%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #29max of 52673.7%
  2. #58none of 52666.3%

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

knowledge

GPQA Diamond

539 results
  1. #75max of 53989.5%
  2. #302none of 53968.6%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #47max of 12285.6%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #56max of 9440.8%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #43max of 9683.5%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #38max of 12286.7%

Leader: Fable 5.1 · 92.4%

long-context

AA-LCR v1.1

523 results
  1. #111max of 52378.3%
  2. #332none of 52342.3%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #72max of 947.2%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #37max of 11559.2%
  2. #45low of 11554.7%
  3. #56none of 11542.5%

Leader: GPT-6.1 Sol · 93.7%

reasoning

ARC-AGI-1

215 results
  1. #112max of 21577.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #123max of 21622.8%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #56max of 52920.9%
  2. #175none of 5293.1%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #40max of 13584.1%

Leader: Fable 5 · 88.6%

reliability

Tool-call error rate

172 results
  1. #53max of 1720.7%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #98max of 18299.3%

Leader: GPT-5 Pro · 100.0%

safety

SRE Bench

32 results
  1. #29max of 320.0%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

351 results
  1. #95none of 3511.1s
  2. #202max of 3513.0s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #80none of 351161.1
  2. #201max of 35182.7

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #52max of 1780.9s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #47max of 17684.5

Leader: Muse Spark 1.1 · 204.5