The LLM benchmark aggregator.

GLM-5.1

Z.ai · 754B · open weights

Strongest in Mathematics (#30 of 140, on 1 of its 2 boards), weakest in Jailbreak resistance (#167 of 272). Above par in 17 of 22 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5 and Claude Opus 5.5 and ahead of O4 Mini and Claude 3.7 Sonnet.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Knowledge55.9−7.6#39/1381/2

Academic knowledge57.1−6.8#36/1231/1

Reasoning54.9−12#43/1782/4

Mathematics56.8−8.9#30/1401/2

Reasoning52.2−16.9#52/1401/3

Science55.9−7.6#52/1221/1

Human preference63.3−4.2#43/3421/1

Agents54.2−13.9#71/2682/5

Knowledge work55.5−18.1#48/1782/2

Coding49−20.1#82/1651/5

Agentic coding48.9−19.8#76/1571/4

Professional48.5−13.1#95/1681/1

Finance52.5−10.4#72/1511/1

Medical47−19.8#97/1401/1

Legal45.9−21.2#99/1511/1

Safety53.2−8.3#97/3371/3

Secure code60.8−5.7#60/2741/1

Fairness50.9−19.1#102/3001/2

Harm refusal52.6−9#117/3001/2

Toxicity avoidance52.7−6.3#158/2721/1

Jailbreak resistance51.5−13.4#167/2721/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for GLM-5.1, left for the other.

Mistral Medium 3.5won 15 · lost 0

Mimo V2.5won 15 · lost 0

Gemini 3.1 Flash Litewon 16 · lost 0

Mistral Largewon 16 · lost 1

Mistral Smallwon 16 · lost 1

Command R Pluswon 15 · lost 1

Gemini 1.5 Flashwon 15 · lost 1

GPT-5.4 Miniwon 15 · lost 1

Jamba 1.6 Miniwon 14 · lost 1

Jamba 1.6 Largewon 14 · lost 1

Llama 4 Scout Instructwon 18 · lost 2

Ling 3.0 Flashwon 16 · lost 2

Gemini 1.5 Prowon 16 · lost 2

Gemma 4 31Bwon 16 · lost 2

GPT-5.4 Nanowon 15 · lost 2

GLM-4.7won 14 · lost 2

GPT-4.1 Nanowon 19 · lost 3

DeepSeek V3.2won 19 · lost 3

Grok 4.3won 19 · lost 3

GPT-4o Miniwon 18 · lost 3

Gemini 2.0 Flashwon 17 · lost 3

Mistral Large 3won 14 · lost 3

Llama 4 Maverick Instructwon 18 · lost 4

DeepSeek V3won 18 · lost 4

Gemini 3.5 Flash Litewon 13 · lost 3

Mimo V2.5 Prowon 17 · lost 4

Gemini 2.5 Flash Litewon 12 · lost 3

Command Awon 16 · lost 4

GPT OSS 20Bwon 15 · lost 4

Llama 3.3 Instruct Turbo (70B)won 11 · lost 3

MiniMax M2.5won 11 · lost 3

Grok 3won 14 · lost 4

DeepSeek R1won 16 · lost 5

GLM-4.6won 16 · lost 5

Grok 4.1 Fastwon 16 · lost 5

Inkling Smallwon 12 · lost 4

Qwen3.6 Pluswon 12 · lost 4

Claude 3.5 Haikuwon 15 · lost 5

Grok 4 Fastwon 11 · lost 4

GLM-5won 11 · lost 4

GPT OSS 120Bwon 16 · lost 6

GPT-4.1 Miniwon 16 · lost 6

Kimi K2 Thinkingwon 13 · lost 5

O3 Miniwon 15 · lost 7

Claude Haiku 4.5won 15 · lost 7

Gemini 2.5 Prowon 15 · lost 7

Kimi K2.5won 15 · lost 7

Gemini 2.5 Flashwon 14 · lost 7

GLM-4.5won 11 · lost 6

Qwen3.5 Flashwon 9 · lost 5

MiniMax M3won 14 · lost 8

GPT-5 Nanowon 12 · lost 7

Mimo V2.6 Flashwon 9 · lost 6

GPT-5 Miniwon 13 · lost 9

Claude Sonnet 4won 13 · lost 9

MiniMax M2.1won 8 · lost 6

Grok 4won 12 · lost 9

MiniMax M2.7won 12 · lost 9

Kimi K2won 12 · lost 9

GPT-4owon 11 · lost 9

DeepSeek V4 Flashwon 12 · lost 10

Kimi K2.6won 12 · lost 10

Claude 3.5 Sonnetwon 10 · lost 9

Claude 3.7 Sonnetwon 11 · lost 10

O4 Miniwon 11 · lost 10

O1 Miniwon 7 · lost 7

Qwen3.8 27Bwon 8 · lost 8

Qwen3 235B A22B Instructwon 10 · lost 10

Claude Opus 4won 10 · lost 10

Inklingwon 11 · lost 11

GPT-5.1won 11 · lost 11

GPT-4.1won 10 · lost 11

O3won 10 · lost 11

GPT-6 Lunawon 9 · lost 10

Claude Sonnet 4.5won 10 · lost 12

GLM-5.2won 10 · lost 12

GPT-6 Solwon 8 · lost 11

GLM-5.3 Flashwon 9 · lost 13

Qwen3.7 Maxwon 9 · lost 13

O1won 8 · lost 12

Claude Opus 4.5won 6 · lost 10

GPT-5won 8 · lost 14

GPT-5.6 Lunawon 8 · lost 14

DeepSeek V4 Prowon 8 · lost 14

Gemini 3.5 Flashwon 8 · lost 14

Gemini 3.6 Flashwon 8 · lost 14

GPT-4 Turbowon 5 · lost 9

Mimo V2.6 Prowon 6 · lost 11

Gemini 3.1 Prowon 6 · lost 11

Muse Sparkwon 4 · lost 11

Grok 4.7won 5 · lost 14

GLM-5.3won 4 · lost 12

Claude Sonnet 4.6won 5 · lost 15

GPT-5.2won 5 · lost 16

Qwen3.8 Maxwon 5 · lost 17

Gemini 3 Prowon 4 · lost 14

Claude Opus 4.6won 4 · lost 14

Gemini 3 Flashwon 3 · lost 12

DeepSeek V4.1 Flashwon 3 · lost 12

Muse Spark 1.1won 4 · lost 18

Gemini 3.8 Flashwon 4 · lost 18

Grok 4.6won 3 · lost 19

Grok 4.5won 3 · lost 19

GPT-5.6 Terrawon 3 · lost 19

Claude Sonnet 5won 2 · lost 20

GPT-5.6 Solwon 2 · lost 20

GPT-6 Astrawon 1 · lost 13

Muse Spark 1.2won 1 · lost 14

Gemini 3.7 Flashwon 1 · lost 15

Kimi K3won 1 · lost 15

GPT-5.4won 1 · lost 21

Claude Opus 4.8won 1 · lost 21

GPT-5.5won 1 · lost 21

Claude Opus 4.7won 1 · lost 21

Claude Fable 5.1won 0 · lost 16

Claude Fable 5won 0 · lost 16

Claude Opus 5.5won 0 · lost 19

Claude Opus 5won 0 · lost 22

§ 3 · Sources

Where the numbers come from

6 publications, 25 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1465

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA 1103

AA-Briefcase ↗1 measureread 2026-09-26

AA-Briefcase 963

SimpleBench ↗1 measureread 2026-09-26

SimpleBench 55.1%

Vals.ai ↗15 measuresread 2026-09-26

Vals · Legal Research Bench 27.89%Vals · CaseLaw 51.55%Vals · LegalBench 84.39%Vals · Harvey Legal Agent Benchmark 0%Vals · Finance Agent 44.79%Vals · CorpFin 64.45%Vals · TaxEval 71.19%Vals · MedCode 41.6%Vals · MedScribe 72.27%Vals · SWE-bench Verified 76.4% (Mini-SWE-agent)Vals · Vibe Code Bench 31.46% (OpenHands)Vals · Code Migration 25.77%Vals · GPQA Diamond 84.52%Vals · MMLU Pro 86.9%Vals · AIME 91.88%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26

Enkrypt · Jailbreak risk 11.5%Enkrypt · Harmful content risk 8.9%Enkrypt · CBRN risk 21%Enkrypt · Toxicity risk 4.4%Enkrypt · Bias risk 76.5%Enkrypt · Insecure code risk 11.6%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=glm-5-1)](https://publicai.io/model-index/m/glm-5-1)