priorsBuilder

Zhipu · released 2026-08-18

GLM 5.3

Measured on 84 benchmarks across 3 reasoning efforts.

General frontier

56.3

#11 of 32 models · best at minimal

Reasoning efforts in your mix

minimal56.3#38 of 128
max49.2#53 of 128
low38.6#92 of 128

Your mix’s benchmarks

7 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #68max of 58742.3%
  2. #107low of 58736.6%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #82max of 52614.3
  2. #175low of 526-8.4

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #24minimal of 19672.0%
  2. #33max of 19666.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #16max of 7069.0%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #35max of 7029.3%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #24max of 9766.2%

Leader: Opus 5.5 · 88.4%

agentic

Terminal-Bench 4.0

35 results
  1. #20max of 3541.8%

Leader: Opus 5.5 · 64.8%

All 84 benchmarks

show 77 more

agentic

AA-Briefcase v1.1

224 results
  1. #31max of 2241509
  2. #88low of 2241187

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #25max of 502.1%

Leader: Opus 5.5 · 15.5%

agentic

AutomationBench-AA

219 results
  1. #24max of 21962.2%
  2. #71low of 21952.0%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #34max of 5036.4%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #26max of 798.3%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #12max of 4546.1%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #9max of 7849.0%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #24max of 3447.5%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #8max of 6939.7%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #12max of 4738.9%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #21max of 405.7%

Leader: GPT-6 Astra · 62.9%

agentic

Vending-Bench 2

67 results
  1. #12max of 67$8,164

Leader: GPT-6 Astra · $15,515

agentic

Vending-Bench Arena

12 results
  1. #4max of 12$7,800

Leader: GPT-6 Astra · $12,400

agentic

τ³-Banking

206 results
  1. #4max of 20650.3%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #21max of 1221622

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #24max of 7744.2%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierCode 1.1 Extended

130 results
  1. #77max of 13052.6%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #69max of 13040.1%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v1

17 results
  1. #2max of 1778.1%

Leader: Fable 5 · 88.2%

coding

FrontierSWE v2

21 results
  1. #10max of 2130.2%

Leader: GPT-6 Astra · 65.5%

coding

IOI

44 results
  1. #13max of 4468.4%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #68max of 12780.5%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #17max of 6216.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #13max of 621.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #28max of 6266.3%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #16max of 22659.0%
  2. #158low of 22642.0%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #6max of 8595.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #26max of 7471.5%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Science-AA

46 results
  1. #26max of 469.5%

Leader: GPT-6 Astra · 63.3%

coding

Terminal-Bench v2.1-AA

238 results
  1. #34max of 23883.9%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #29max of 21641.9%
  2. #35low of 21634.8%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench 1-100

22 results
  1. #9max of 2220.0%

Leader: Opus 5.5 · 30.4%

coding

Vibe Code Bench v1.1

105 results
  1. #29max of 10578.1%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #25max of 12775.4%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #38max of 61044.8
  2. #91low of 61034.3

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #26max of 2201478

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #38max of 2201481

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #19max of 4753.5%

Leader: Gemini 4 · 68.9%

composite

Vals RSI Index v1.1

24 results
  1. #12max of 2422.9%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #8max of 2417.4%

Leader: GPT-6 Sol · 22.8%

composite

Vals RSI Index v1.1: Post-training: Finance Agent

24 results
  1. #9max of 240.0%

Leader: Opus 5.5 · 35.1%

composite

Vals RSI Index v1.1: Pre-training: Compression

24 results
  1. #13max of 2433.4%

Leader: Opus 5.5 · 46.5%

composite

Vals RSI Index v1.1: Pre-training: LM Training

24 results
  1. #12max of 2440.8%

Leader: Opus 5.5 · 53.0%

cost

AA Intelligence Index Cost per Task

177 results
  1. #106low of 177$0.85
  2. #148max of 177$2.01

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #96low of 177$1,013
  2. #146max of 177$2,503

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #19max of 721799

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (text)

55 results
  1. #18max of 551786

Leader: GPT-6 Astra · 2732

economics

Excel Modeling Benchmark

74 results
  1. #45max of 7456.3%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #20max of 7855.8%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #115max of 21011.2%
  2. #136low of 2108.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #17max of 2881651
  2. #95low of 2881322

Leader: Opus 5.5 · 1866

economics

TaxEval (Vals)

125 results
  1. #62max of 12572.4%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #158low of 52633.9%
  2. #159max of 52633.9%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #39max of 52670.4%
  2. #178low of 52636.0%

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

knowledge

GPQA Diamond

539 results
  1. #45max of 53991.7%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #36max of 12288.1%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

Harvey LAB-AA v1.1

27 results
  1. #21max of 270.3%

Leader: Muse Wonder Full (Max) · 13.1%

knowledge

MedCode

94 results
  1. #46max of 9442.9%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #10max of 9688.8%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #37max of 12286.8%

Leader: Fable 5.1 · 92.4%

knowledge

Public Benefits Bench v1.1

50 results
  1. #10max of 5068.5%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #89max of 52379.7%
  2. #185low of 52372.7%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #13max of 9448.3%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #25max of 11568.8%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

52 results
  1. #33max of 5249.0%

Leader: Sonnet 5.5 · 100.0%

reasoning

CritPt

529 results
  1. #63max of 52919.1%
  2. #92low of 52914.6%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #28max of 13584.8%

Leader: Fable 5 · 88.6%

reasoning

MysteryMechanism

27 results
  1. #18max of 2723.0%

Leader: GPT-6 Astra · 53.2%

reliability

Tool-call error rate

171 results
  1. #61max of 1710.8%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #116max of 18198.9%

Leader: GPT-5 Pro · 100.0%

safety

CWE-Bench

19 results
  1. #9max of 1955.8%

Leader: Grok 4.7 · 68.3%

safety

CyberBench v1.1

45 results
  1. #19max of 4572.1%

Leader: Wonder · 80.4%

safety

SRE Bench

32 results
  1. #13max of 329.2%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

351 results
  1. #200low of 3512.9s
  2. #212max of 3513.3s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #220low of 35174.2
  2. #241max of 35166.9

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #70max of 1771.1s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #60max of 17679.5

Leader: Muse Spark 1.1 · 204.3