The LLM benchmark aggregator.

GLM-4.6

Z.ai · 357B · open weights

Strongest in Tool use (#4 of 81), weakest in Harm refusal (#264 of 300). Above par in 15 of 25 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5 and Claude Opus 5.5 and ahead of GLM-4.5 and Gemini 3.5 Flash Lite.

§ 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.

Agents59.5−8.6#36/2682/5

Tool use74leads#4/811/1

Knowledge work50.5−23.1#78/1781/2

Reasoning53−13.9#64/1781/4

Mathematics57−8.7#27/1401/2

Science48.9−14.6#80/1221/1

Professional50.4−11.2#76/1681/1

Medical54.7−12.1#44/1401/1

Legal51.1−16#69/1511/1

Finance47.5−15.4#102/1511/1

Knowledge51−12.5#81/1381/2

Academic knowledge51.2−12.7#75/1231/1

Human preference59.3−8.2#98/3421/1

Core abilities47.6−19.8#119/2041/3

General intelligence46.6−23.5#125/2041/3

Coding43.3−25.8#129/1651/5

Agentic coding41.9−26.8#130/1571/4

Safety47.1−14.4#254/3372/3

Factual grounding51.2−19.4#51/1011/1

Secure code56−10.5#110/2741/1

Toxicity avoidance52.4−6.6#168/2721/1

Jailbreak resistance40.6−24.3#216/2721/1

Fairness43.9−26.1#223/3001/2

Harm refusal43.6−18#264/3001/2

§ 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-4.6, left for the other.

Mistral Medium 3.5won 15 · lost 2

Jamba 1.6 Miniwon 13 · lost 2

Mistral Large 3won 16 · lost 3

Command Awon 20 · lost 4

Command R Pluswon 13 · lost 3

GPT OSS 20Bwon 17 · lost 4

Claude Haiku 4.5won 20 · lost 5

Ling 3.0 Flashwon 14 · lost 4

Mistral Smallwon 16 · lost 5

GPT-4.1 Nanowon 18 · lost 6

Llama 4 Scout Instructwon 17 · lost 6

Mistral Largewon 14 · lost 5

Gemini 2.5 Flash Litewon 14 · lost 5

GPT-5.4 Nanowon 14 · lost 5

Jamba 1.6 Largewon 11 · lost 4

GPT-4o Miniwon 16 · lost 6

Gemini 2.0 Flashwon 15 · lost 6

GPT OSS 120Bwon 17 · lost 7

Gemini 1.5 Prowon 12 · lost 5

Gemini 1.5 Flashwon 11 · lost 5

Llama 3.3 Instruct Turbo (70B)won 10 · lost 5

Claude 3.5 Haikuwon 13 · lost 7

Grok 4 Fastwon 11 · lost 6

GPT-5 Nanowon 14 · lost 8

Gemma 3 27B Itwon 10 · lost 6

DeepSeek V3won 15 · lost 9

Kimi K2won 15 · lost 9

DeepSeek R1won 14 · lost 9

Nemotron 3 Ultrawon 9 · lost 6

Grok 3won 12 · lost 8

Llama 4 Maverick Instructwon 15 · lost 10

O4 Miniwon 14 · lost 10

O3 Miniwon 12 · lost 9

GPT-4owon 11 · lost 9

Grok 4.1 Fastwon 12 · lost 10

GPT-4.1 Miniwon 13 · lost 11

MiniMax M2.5won 8 · lost 7

Gemini 3.5 Flash Litewon 9 · lost 8

GLM-4.5won 10 · lost 9

Claude 3.5 Sonnetwon 9 · lost 9

Qwen3 235B A22B Instructwon 11 · lost 12

Grok 4.3won 11 · lost 12

MiniMax M2.7won 10 · lost 11

Qwen3.6 Pluswon 8 · lost 9

Qwen3.5 Flashwon 7 · lost 8

Inkling Smallwon 7 · lost 8

GPT-4.1won 11 · lost 13

GPT-5.4 Miniwon 8 · lost 10

GPT-5 Miniwon 11 · lost 14

Grok 4won 10 · lost 13

Gemini 3.1 Flash Litewon 8 · lost 11

Kimi K2.6won 10 · lost 14

Qwen3 Maxwon 6 · lost 9

Mimo V2.5won 6 · lost 9

Mimo V2.6 Flashwon 6 · lost 9

DeepSeek V3.2won 10 · lost 15

Claude Sonnet 4won 9 · lost 15

Gemini 2.5 Flashwon 9 · lost 15

Gemma 4 31Bwon 8 · lost 14

Claude 3.7 Sonnetwon 7 · lost 13

GPT-6 Lunawon 7 · lost 13

GLM-4.7won 5 · lost 10

GLM-5.2won 8 · lost 16

Mimo V2.5 Prowon 7 · lost 16

MiniMax M3won 7 · lost 17

Inklingwon 7 · lost 17

Kimi K2.5won 7 · lost 17

Claude Sonnet 4.6won 6 · lost 16

O1won 5 · lost 14

Gemini 3.1 Prowon 5 · lost 14

DeepSeek V4 Flashwon 6 · lost 17

Qwen3.7 Maxwon 6 · lost 17

DeepSeek V4.1 Flashwon 4 · lost 12

Gemini 2.5 Prowon 6 · lost 18

GPT-5.6 Lunawon 6 · lost 18

Claude Sonnet 4.5won 6 · lost 19

GLM-5.1won 5 · lost 16

GLM-5won 4 · lost 13

Claude Opus 4won 5 · lost 17

Qwen3.8 27Bwon 4 · lost 14

O3won 5 · lost 18

GLM-5.3 Flashwon 5 · lost 18

Gemini 3.5 Flashwon 5 · lost 18

Gemini 3.6 Flashwon 5 · lost 18

Kimi K2 Thinkingwon 4 · lost 15

DeepSeek V4 Prowon 4 · lost 20

Claude Opus 4.6won 3 · lost 17

Gemini 3 Flashwon 2 · lost 13

Muse Sparkwon 2 · lost 13

Grok 4.5won 3 · lost 20

Qwen3.8 Maxwon 3 · lost 20

GPT-5.2won 3 · lost 21

GPT-5.6 Terrawon 3 · lost 21

GPT-5.6 Solwon 3 · lost 21

Gemini 3.7 Flashwon 2 · lost 15

Claude Opus 4.5won 2 · lost 17

Mimo V2.6 Prowon 2 · lost 17

Grok 4.7won 2 · lost 18

GPT-6 Solwon 2 · lost 19

Gemini 3 Prowon 2 · lost 19

GPT-5.1won 2 · lost 20

Muse Spark 1.1won 2 · lost 21

Gemini 3.8 Flashwon 2 · lost 21

GPT-5won 2 · lost 22

Claude Sonnet 5won 2 · lost 22

GPT-5.4won 2 · lost 22

GPT-5.5won 2 · lost 22

Claude Opus 4.7won 2 · lost 22

Muse Spark 1.2won 1 · lost 15

GLM-5.3won 1 · lost 16

Kimi K3won 1 · lost 16

Grok 4.6won 1 · lost 22

Claude Opus 4.8won 1 · lost 22

GPT-6 Astrawon 0 · lost 16

Claude Fable 5.1won 0 · lost 17

Claude Fable 5won 0 · lost 17

Claude Opus 5.5won 0 · lost 20

Claude Opus 5won 0 · lost 23

§ 3 · Sources

Where the numbers come from

7 publications, 19 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1424

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA 746

BFCL ↗1 measureread 2026-09-26

BFCL v4 72.38%

Kagi LLM Benchmark ↗1 measureread 2026-09-26

Kagi LLM Benchmark 45.7%

Vals.ai ↗8 measuresread 2026-09-26

Vals · LegalBench 79.61%Vals · CorpFin 56.84%Vals · TaxEval 66.23%Vals · MedQA 92.22%Vals · Vibe Code Bench 3.09% (OpenHands)Vals · GPQA Diamond 74.5%Vals · MMLU Pro 82.2%Vals · AIME 92.71%

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

Enkrypt · Jailbreak risk 20.7%Enkrypt · Harmful content risk 45%Enkrypt · CBRN risk 25.3%Enkrypt · Toxicity risk 4.6%Enkrypt · Bias risk 87.3%Enkrypt · Insecure code risk 21.3%

Vectara Hallucination Leaderboard ↗1 measureread 2026-09-26

Vectara · Factual consistency 90.5%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=glm-4-6)](https://publicai.io/model-index/m/glm-4-6)