Benchfolio

Leaderboard / Model

GLM-5.1

Z.ai (Zhipu AI) · Open weights · Released Apr 7, 2026 · $1.40 in · $4.40 out per 1M tokens

32Overall · rank 28

Score by job

Coding

Not enough data · 2 of 5 benchmarks

BenchmarkResultVariantz
SWE-bench Verified74.2%glm-5.1-0.66
LMArena Coding1486 ratingglm-5.1-0.24

Agents

Not enough data · 1 of 4 benchmarks

BenchmarkResultVariantz
APEX-Agents40.9%glm-5.1-1.09

Reasoning

24 · rank 32 · 2 of 4 benchmarks

BenchmarkResultVariantz
GPQA Diamond89.9%glm-5.1-0.67
SimpleBench55.1%glm-5.1-1.03

Math

15 · rank 36 · 4 of 5 benchmarks

BenchmarkResultVariantz
FrontierMath Tiers 1–336.8%glm-5.1-2.15
OTIS Mock AIME93.3%glm-5.1-0.55
ProofBench22.2%glm-5.1-1.01
LMArena Math1473 ratingglm-5.1-0.26

Writing

57 · rank 26 · 2 of 2 benchmarks

BenchmarkResultVariantz
LMArena Creative Writing1452 ratingglm-5.10.03
LMArena Instruction Following1450 ratingglm-5.1-0.53

Design

30 · rank 33 · 1 of 1 benchmarks

BenchmarkResultVariantz
LMArena WebDev1508 ratingglm-5.1-0.49