Model profile / Z.AI

GLM-5.3

Z.AI's GLM-5.3 flagship uses the GLM-5.2 base with additional post-training for coding, long-horizon agents and cyber evaluation. Thinking is always on, with low, high and max effort.

14 Aug 2026Open weights announcedProvider

Sources & details

Four areas of evidence

The capabilities behind the model.

How the indices work

Reasoning

115.3Index points

Mathematical and scientific problem solving.

This model│ Other models · 155 rated

Score and research range

Full-precision point: 115.27800041779444
Range: 110.75–119.81 · 90% Research Range
Scope: R0 latent score conditional on frozen working model and native aggregate portfolio

Score uses the admitted direct native evidence under the selected method.

Coding

107.8Index points

Software engineering and programming.

This model│ Other models · 39 rated

Score and research range

Full-precision point: 107.83687748877978
Range: No qualified range available
Scope: Frozen Coding score under admitted comparison/reference/source reweighting

Score uses the admitted direct native evidence under the selected method.

General

54.4Index points

Factual reliability and instruction following.

This model│ Other models · 61 rated

Score and research range

Full-precision point: 54.423555555555566
Range: No qualified range available
Scope: General G4 under admitted instruction-task portfolio resampling conditional on exact recorded R and U

Score uses the admitted direct native evidence under the selected method.

Agentic

113.6Index points

Tool use, workflows and professional tasks.

This model│ Other models · 40 rated

Score and research range

Full-precision point: 113.56121206143716
Range: 110.72–117.24 · 90% Research Range
Scope: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes

Score uses the admitted direct native evidence under the selected method.

26retained native records

18evidence families

Capabilities estimateQualified original evidence across four areas.

Current index snapshot · 2026-09-16. Each category uses its own scale; scores across categories are not directly comparable. Evidence counts describe the retained Capabilities inputs; a family can contain several tasks or configurations. A point lead does not establish superiority.

Performance in context

Capabilities, price & speed.

Explore every model

GLM-5.3 is highlighted wherever a compatible measurement is available.

Lumina Capabilities Index

Capabilities versus API price

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 62 models · 1 highlighted

Chart loads as you explore

Lumina Capabilities Index

Capabilities versus output speed

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 72 models · Developer API · 10K · 1 highlighted

Chart loads as you explore

Explore the detail

Sources & model details.

Original evidence, with its context.

Current index evidence 5Scores, research ranges and the sources behind each category.

Capabilities

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: C1 native factor under conditional crossed source/family reweighting of admitted fixed-corpus reference and own portfolio
Sources: Artificial Analysis:Artificial Analysis · Datacurve:Datacurve · Epoch AI:Epoch AI · HKUST-NLP:Z.ai · LiveBench:LiveBench · UNKNOWN:Cognition · UNKNOWN:Terminal-Bench owners
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Coding

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Frozen Coding score under admitted comparison/reference/source reweighting
Sources: Artificial Analysis · Cognition · Datacurve · LiveBench
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Reasoning

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: R0 latent score conditional on frozen working model and native aggregate portfolio
Sources: aa · epoch
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Agentic

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes
Sources: Artificial Analysis · Toolathlon · owner
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

General

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: General G4 under admitted instruction-task portfolio resampling conditional on exact recorded R and U
Sources: Artificial Analysis · LiveBench
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Benchmark measurements 30Published values in each benchmark’s own units.

30 measurements

GLM-5.3 published benchmark measurements
Benchmark	Result	Published configuration	Source
AA Long Context Reasoningstandard	76.333%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis AA-LCR evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
CritPtstandard	19.143%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis CritPt evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
GDPval-AA v2v2	63.149%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis GDPval-AA v2 using Stirrup	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
GPQA Diamonddiamond	91.717%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis GPQA Diamond evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
Humanity's Last Examtext-only current	42.261%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis text-only Humanity's Last Exam evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
SciCoderolling	56.481%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis SciCode evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · reference-only
τ³-Banking3	50.309%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis tau3-Banking evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
Terminal-Bench2.1	83.895%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis Terminal-Bench v2.1 in e2b	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
FrontierMath v2 Tiers 1-3v2 / task implementation 2.0.0	68.772%	GLM-5.3 (max)Epoch AI benchmark runner	Epoch AI permanent refresh source ↗Observed 2026-08-25 · checked 2026-09-01independently-verified · reference-only
FrontierMath v2 Tier 4v2 / task implementation 2.0.0	29.268%	GLM-5.3 (max)Epoch AI benchmark runner	Epoch AI permanent refresh source ↗Observed 2026-08-25 · checked 2026-09-01independently-verified · reference-only

Showing 10 of 30 measurements

Specifications & source recordsDocumented model details and original references.

Model ID / checkpoint: glm-5.3
Weight access: Open weights announced
Context window: 1,000,000
Maximum output: 128,000
Modalities: text
Details checked: 31 Aug 2026

GLM-5.3 individual evaluationsArtificial Analysis · Published 18 Aug 2026 · checked 27 Aug 2026 · source-checked Z.AI model pricingZ.AI · checked 31 Aug 2026 · source-checked GLM-5.3: Frontier Coding with Emergent Cyber CapabilitiesZ.AI · Published 14 Aug 2026 · checked 15 Aug 2026 · source-checked GLM-5.3 developer overviewZ.AI · checked 31 Aug 2026 · source-checked Epoch AI permanent refresh sourceEpoch AI · checked 1 Sept 2026 · source-checked

Specialist task evidenceSeparate agent runs and execution conditions.

All models

Model profile / Z.AI

GLM-5.3

Z.AI's GLM-5.3 flagship uses the GLM-5.2 base with additional post-training for coding, long-horizon agents and cyber evaluation. Thinking is always on, with low, high and max effort.

14 Aug 2026Open weights announcedProvider

Sources & details

Four areas of evidence

The capabilities behind the model.

How the indices work

Reasoning

115.3Index points

Mathematical and scientific problem solving.

This model│ Other models · 155 rated

Score and research range

Full-precision point: 115.27800041779444
Range: 110.75–119.81 · 90% Research Range
Scope: R0 latent score conditional on frozen working model and native aggregate portfolio

Score uses the admitted direct native evidence under the selected method.

Coding

107.8Index points

Software engineering and programming.

This model│ Other models · 39 rated

Score and research range

Full-precision point: 107.83687748877978
Range: No qualified range available
Scope: Frozen Coding score under admitted comparison/reference/source reweighting

Score uses the admitted direct native evidence under the selected method.

General

54.4Index points

Factual reliability and instruction following.

This model│ Other models · 61 rated

Score and research range

Full-precision point: 54.423555555555566
Range: No qualified range available
Scope: General G4 under admitted instruction-task portfolio resampling conditional on exact recorded R and U

Score uses the admitted direct native evidence under the selected method.

Agentic

113.6Index points

Tool use, workflows and professional tasks.

This model│ Other models · 40 rated

Score and research range

Full-precision point: 113.56121206143716
Range: 110.72–117.24 · 90% Research Range
Scope: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes

Score uses the admitted direct native evidence under the selected method.

26retained native records

18evidence families

Capabilities estimateQualified original evidence across four areas.

Performance in context

Capabilities, price & speed.

Explore every model

GLM-5.3 is highlighted wherever a compatible measurement is available.

Lumina Capabilities Index

Capabilities versus API price

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 62 models · 1 highlighted

Chart loads as you explore

Lumina Capabilities Index

Capabilities versus output speed

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 72 models · Developer API · 10K · 1 highlighted

Chart loads as you explore

Explore the detail

Sources & model details.

Original evidence, with its context.

Current index evidence 5Scores, research ranges and the sources behind each category.

Capabilities

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: C1 native factor under conditional crossed source/family reweighting of admitted fixed-corpus reference and own portfolio
Sources: Artificial Analysis:Artificial Analysis · Datacurve:Datacurve · Epoch AI:Epoch AI · HKUST-NLP:Z.ai · LiveBench:LiveBench · UNKNOWN:Cognition · UNKNOWN:Terminal-Bench owners
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Coding

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Frozen Coding score under admitted comparison/reference/source reweighting
Sources: Artificial Analysis · Cognition · Datacurve · LiveBench
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Reasoning

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: R0 latent score conditional on frozen working model and native aggregate portfolio
Sources: aa · epoch
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Agentic

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes
Sources: Artificial Analysis · Toolathlon · owner
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

General

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: glm-5-3
Method: methods-a4471cee84fd478d2ed3
Range interpretation: General G4 under admitted instruction-task portfolio resampling conditional on exact recorded R and U
Sources: Artificial Analysis · LiveBench
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Benchmark measurements 30Published values in each benchmark’s own units.

30 measurements

GLM-5.3 published benchmark measurements
Benchmark	Result	Published configuration	Source
AA Long Context Reasoningstandard	76.333%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis AA-LCR evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
CritPtstandard	19.143%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis CritPt evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
GDPval-AA v2v2	63.149%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis GDPval-AA v2 using Stirrup	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
GPQA Diamonddiamond	91.717%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis GPQA Diamond evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
Humanity's Last Examtext-only current	42.261%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis text-only Humanity's Last Exam evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
SciCoderolling	56.481%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis SciCode evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · reference-only
τ³-Banking3	50.309%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis tau3-Banking evaluation	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
Terminal-Bench2.1	83.895%	GLM-5.3 (max; Artificial Analysis independent run)Artificial Analysis Terminal-Bench v2.1 in e2b	GLM-5.3 individual evaluations ↗Observed 2026-08-27 · checked 2026-08-27independently-verified · ranking-eligible
FrontierMath v2 Tiers 1-3v2 / task implementation 2.0.0	68.772%	GLM-5.3 (max)Epoch AI benchmark runner	Epoch AI permanent refresh source ↗Observed 2026-08-25 · checked 2026-09-01independently-verified · reference-only
FrontierMath v2 Tier 4v2 / task implementation 2.0.0	29.268%	GLM-5.3 (max)Epoch AI benchmark runner	Epoch AI permanent refresh source ↗Observed 2026-08-25 · checked 2026-09-01independently-verified · reference-only

Showing 10 of 30 measurements

Specifications & source recordsDocumented model details and original references.

Model ID / checkpoint: glm-5.3
Weight access: Open weights announced
Context window: 1,000,000
Maximum output: 128,000
Modalities: text
Details checked: 31 Aug 2026

Specialist task evidenceSeparate agent runs and execution conditions.

The capabilities behind the model.

Capabilities, price & speed.

Capabilities versus API price

Capabilities versus output speed

Sources & model details.

Capabilities

Coding

Reasoning

Agentic

General

More from Z.AI.

GLM-5.3-Flash

GLM-5.2

GLM-5.1

The capabilities behind the model.

Capabilities, price & speed.

Capabilities versus API price

Capabilities versus output speed

Sources & model details.

Capabilities

Coding

Reasoning

Agentic

General

More from Z.AI.

GLM-5.3-Flash

GLM-5.2

GLM-5.1