GLM-5.2

Across the task benchmarks we track, GLM-5.2’s strongest relative results are in terminal use, knowledge work, and work automations. On Terminal-Bench 2.1, it scores 77.9%, ranking #5 among 13 models with reported results in our dataset.

Updated: Nov 5, 2026Data as of: Nov 2, 2026

Compare this model

Strongest results

Strongest results for GLM-5.2 in tracked evaluation cohorts
Benchmark	Measures	Score
Humanity's Last Exam ↗Stable HLE — text-only subset	Expert reasoning	41.15%#29 of 68
AA-Briefcase ↗1.1 / combined Elo	Business deliverables	1,227.23 Elo#34 of 57
GDPval-AA ↗2.1	Knowledge work	1,381.49 Elo#25 of 38
IFBench ↗58 constraint types	Precise instruction following	73.33%#12 of 17

Model details

Open source / open weights10 benchmarks

Specifications for GLM-5.2
Model	GLM-5.2
Agent Index v1.2	Insufficient coverage (3/7) · Preview
Creator	Z.ai (opens in a new tab)
Availability	Open source / open weights
Context size	1,000,000 tokens (opens in a new tab)
Max output	Not reported
Knowledge cutoff	Not reported
License	MIT · Model weights ↗ (opens in a new tab)

Pricing and performance

Per provider and reported configuration. Prices are USD per 1M tokens.

Provider offerings for GLM-5.2
Provider / configuration	Contexttokens	InputUSD / 1M tokens	OutputUSD / 1M tokens	LatencyTTFT · seconds	Speedtokens / sec
Z.ai (opens in a new tab) GLM-5.2 (Max)	1,000,000	$1.40 (opens in a new tab)	$4.40 (opens in a new tab)	1.777 (opens in a new tab)	94.073 (opens in a new tab)

Benchmark performance

GLM-5.2 in color, alongside leading models. Latest stored reports, best effort per report date.

Reasoning

198-question Diamond subset; independent AA evaluation

GPQA Diamond ↗

01
GPT-6 Astra96.3%
02
Gemini 3.8 Flash95.3%
03
Gemini 3.1 Pro94.1%
03
GPT-5.6 Sol94.1%
03
Muse Spark 1.394.1%
23
GLM-5.2This model89.5%

Terminal Use

89 tasks; Terminus 2; e2b; three repeats

Terminal-Bench 2.1 ↗

01
Kimi K385%
02
GLM-5.3-Flash84.3%
03
GLM-5.383.9%
04
DeepSeek V4 Pro 081378.7%
05
GLM-5.2This model77.9%

Work Automations

Private 657 tasks; 50 turns; objective completion

AutomationBench-AA ↗

01
Gemini 4 Carbon77.8%
02
Gemini 4 Argon77.5%
03
Kimi K3.1 Pro73.3%
04
MiniMax M3.1 Flash72.52%
05
Claude Sonnet 5.571.8%
40
GLM-5.2This model28.4%

Knowledge Work

220 tasks; Stirrup agent; pairwise Elo

GDPval-AA v2.1 ↗

01
Claude Opus 5.51,866 Elo
02
Kimi K3.1 Pro1,858 Elo
03
Claude Sonnet 5.51,838 Elo
04
Claude Fable 5.11,763 Elo
05
DeepSeek V4.1 Flash 10171,746 Elo
25
GLM-5.2This model1,381 Elo

Reported benchmarks

All published results in our dataset, with versions, evaluation settings, and sources.

All published benchmark reports for GLM-5.2
Benchmark	Score	Evaluation setup	Source
AA-Briefcase1.1 / combined EloCombined Elo · Higher is better	1,227.23 Elo	selection: best reported configuration per release · GLM-5.2 (Max) · tools: benchmark-defined	Artificial Analysis (opens in a new tab)independent
AutomationBench-AA1.0.6 / AA private 657Objectives completed without guardrail violations · Higher is better	28.399%	GLM-5.2 (Max); Private 657 tasks; 50 turns; objective completion	Artificial Analysis (opens in a new tab)independent
GDP.pdf100 tasks / All-pass / five attemptsAccuracy · Higher is better	10.4%	selection: best reported configuration per release · GLM-5.2 (Max) · tools: benchmark-defined	Artificial Analysis (opens in a new tab)independent
GDPval-AA2.1Elo · Higher is better	1,381.49 Elo	GLM-5.2 (Max); 220 tasks; Stirrup agent; pairwise Elo	Artificial Analysis (opens in a new tab)independent
GDPval-AA2.1Elo · Higher is better	1,251.81 Elo	GLM-5.2 (Non-reasoning); 220 tasks; Stirrup agent; pairwise Elo	Artificial Analysis (opens in a new tab)independent
GPQADiamond / Artificial AnalysisAccuracy · Higher is better	89.495%	GLM-5.2 (Max); 198-question Diamond subset; independent AA evaluation	Artificial Analysis (opens in a new tab)independent
GPQADiamond / Artificial AnalysisAccuracy · Higher is better	68.586%	GLM-5.2 (Non-reasoning); 198-question Diamond subset; independent AA evaluation	Artificial Analysis (opens in a new tab)independent
Humanity's Last ExamStable HLE — text-only subsetAccuracy · Higher is better	54.7%	Text-only per model-card footnote; 163840 generation; 300K context; no context management; GPT-5.5 medium judge · Tools enabled	Z.ai ↗ (opens in a new tab)provider reported
Humanity's Last ExamStable HLE — text-only subsetAccuracy · Higher is better	41.149%	GLM-5.2 (Max); original HLE text-only 2,158 questions; no tools; not HLE Diamond · No tools	Artificial Analysis (opens in a new tab)independent
Humanity's Last ExamStable HLE — text-only subsetAccuracy · Higher is better	9.778%	GLM-5.2 (Non-reasoning); original HLE text-only 2,158 questions; no tools; not HLE Diamond · No tools	Artificial Analysis (opens in a new tab)independent
IFBench58 constraint typesAccuracy · Higher is better	73.333%	selection: best reported configuration per release · GLM-5.2 (Max) · tools: benchmark-defined	Artificial Analysis (opens in a new tab)independent
Terminal-Bench2.1 / Artificial AnalysisPass@1 · Higher is better	77.903%	GLM-5.2 (Max); 89 tasks; Terminus 2; e2b; three repeats	Artificial Analysis (opens in a new tab)independent
Terminal-Bench2.1 / Artificial AnalysisPass@1 · Higher is better	51.685%	GLM-5.2 (Non-reasoning); 89 tasks; Terminus 2; e2b; three repeats	Artificial Analysis (opens in a new tab)independent
Terminal-Bench2.1 / published open-model reportsTask success · Higher is better	81%	harness comparison: provider-reported; settings differ across models · Terminus 2; JSON parser; 4h; 500 episodes; 256K context	Z.ai ↗ (opens in a new tab)provider reported
Terminal-Bench4.0 / mini-swe-agentTask success · Higher is better	1.01%	harness: mini-swe-agent · repeats: 3 · selection: best reported configuration per release · GLM-5.2 (Max)	Artificial Analysis (opens in a new tab)independent