RankingEffort benchmarks

GLM-5.3-Flash · Reasoning

Compare models

50 measured results · 6 effort settings · 48 benchmark/harness combinations

Model evidence and sources →

Change model

Model

Current configurationSources

Compare with

Choose a second model and its effort to see both profiles and benchmark differences below.

Compare withModel and configuration

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AGLM-5.3-Flash · Reasoning2 of 7 capabilities supported

Gaps are unknown, not zero. Hollow points are preliminary.

Agentic
A50.6Supported
Hard reasoning
A42.5Supported
Coding
A66.4Preliminary
Human pref
Unknown
Knowledge
A17.0Preliminary
Multimodal
Unknown
Long context
A72.3Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Capability score evidence

GLM-5.3-Flash · Reasoning: 40.6 overall comparison score. Meets the overall evidence requirements.

5 areas measured. The available task mix affects this estimate; missing areas remain unknown.

Which sources carry the weight?

Artificial Analysis · 100.0% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

Humanity’s Last Exam · 7.8% direct comparison weight
Critpt · 7.8% direct comparison weight
GPQA · 7.8% direct comparison weight
omniscience · 7.8% direct comparison weight
GDPval · 7.2% direct comparison weight
Other families (11) · 61.5% direct comparison weight

These shares describe direct comparison weight, not fractions of the final score or confidence. Source balancing does not give every family equal weight or remove correlations between overlapping sources.

How sensitive is this score?

The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-10-20.

Both ranks use the same surviving configurations for each removal. The cohort starts with 444 configurations supported in the original audit, including separate effort settings. These are diagnostic positions, not the default leaderboard ranks. Eligibility is not rechecked after removal.

What changes when a source is removed?
Removed source	Rank before → after	Rank movement	Cohort remaining / lost support	Score after (change)
Artificial Analysis	Comparison support lost; no rank		203 remaining / 241 lost	Unavailable
ARC Prize	85 → 86	Down 1	444 remaining / 0 lost	46.3 (+5.7 points)
Datacurve	85 → 88	Down 3	444 remaining / 0 lost	36.1 (-4.5 points)
LiveCodeBench	85 → 85	No movement	444 remaining / 0 lost	40.5 (-0.1 points)
Scale	85 → 84	Up 1	444 remaining / 0 lost	42.8 (+2.2 points)
Terminal-Bench	85 → 85	No movement	444 remaining / 0 lost	40.7 (+0.0 points)
Cognition	85 → 88	Down 3	444 remaining / 0 lost	39.1 (-1.6 points)
LiveBench	85 → 85	No movement	443 remaining / 1 lost	41.5 (+0.9 points)
SWE-rebench	85 → 83	Up 2	444 remaining / 0 lost	40.9 (+0.2 points)
Epoch AI	84 → 87	Down 3	431 remaining / 13 lost	41.2 (+0.6 points)
Vals AI	85 → 84	Up 1	443 remaining / 1 lost	39.8 (-0.9 points)
Provider reports · Anthropic	85 → 85	No movement	444 remaining / 0 lost	40.6 (-0.0 points)
Provider reports · Moonshot AI	85 → 85	No movement	444 remaining / 0 lost	40.0 (-0.6 points)
Provider reports · Meta	85 → 85	No movement	444 remaining / 0 lost	40.6 (-0.1 points)
Provider reports · Mistral	85 → 85	No movement	444 remaining / 0 lost	40.6 (+0.0 points)
Provider reports · OpenAI	85 → 85	No movement	444 remaining / 0 lost	40.8 (+0.2 points)
All provider reports	85 → 85	No movement	444 remaining / 0 lost	40.1 (-0.5 points)

Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.

Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit

Every benchmark family

Removed evidence	Recomputed score	Change
arc-agi	45.4	+4.8
deepswe	36.1	-4.5
omniscience	44.8	+4.1
enterprise-ops	44.3	+3.7
automationbench	38.4	-2.2
itbench	38.6	-2.0
gdpval	38.6	-2.0
frontiercode	39.1	-1.6
briefcase	39.1	-1.5
mmmu-pro	42.1	+1.4
tau-bench	39.3	-1.4
hle	42.0	+1.3
aa-lcr	39.3	-1.3
terminal-bench-science	41.8	+1.2
analyst-agent	41.8	+1.1
terminal-bench	39.5	-1.1
gdp-pdf	41.7	+1.1
simpleqa	41.5	+0.9
epoch-game-puzzles	39.8	-0.9
ifbench	39.9	-0.7
mmlu-pro	41.1	+0.5
critpt	41.0	+0.4
multi-swe-bench	40.2	-0.4
livebench-instructions	41.0	+0.4
harvey	41.0	+0.4
gpqa	40.3	-0.3
livebench-language	40.9	+0.3
swe-atlas-qna	40.4	-0.3
arc-agi-3	40.4	-0.3
swe-rebench	40.9	+0.2
apex-agents	40.8	+0.2
livebench-coding	40.4	-0.2
enigma-eval	40.5	-0.2
vibe-code	40.5	-0.1
aime	40.5	-0.1
livebench-reasoning	40.5	-0.1
swe-atlas-test-writing	40.5	-0.1
livecodebench	40.5	-0.1
vals-legal-research	40.5	-0.1
vals-code-migration	40.6	-0.1
vals-tax-agent	40.6	-0.1
vals-finance-agent	40.6	-0.1
swe-atlas-refactoring	40.6	-0.0
livebench-data	40.6	+0.0
vals-excel-modeling	40.6	-0.0
frontiermath	40.6	+0.0
math500	40.6	+0.0
swe-bench-pro	40.6	+0.0
livebench-math	40.6	-0.0
gmmlu	40.6	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 65 observations, 39 contributing, 0 matched without graph weight and 26 excluded. Counts do not establish rank eligibility.

3 · Benchmark family or source protocol has not been reviewed
1 · Original SciCode grading defects are under review; retained as supporting evidence, excluded from aggregate and capability scores. Corrected SciCode-Verified requires a separate protocol review. Sources: https://arxiv.org/abs/2608.04975 and https://artificialanalysis.ai/evaluations/scicode
3 · Effort is not identified
4 · Evaluator has not been reviewed for independent admission
9 · Supporting evidence outside the reviewed capability core
6 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately

Inspect every observation and exclusion →

Benchmark results

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / source	Low	High	Max	Reasoning	None	Unspecified
AA-AnalystAgentArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of tasks solved on all 5 of 5 attempts · 80 tasks, 5 attempts per task · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	25% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	Not reported	Not reported	Not reported	1,453 Elo Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	80% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
AutomationBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of task objectives completed with no guardrail violations · Higher is better · Benchmark developed by Zapier · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	60.37% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally. CritPt is under review by its authors following external feedback · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	15.43% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
EnterpriseOps-Gym-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Overall task success rate across all eight enterprise domains (oracle tool mode) · Benchmark developed by ServiceNow Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	33.21% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
GDP.pdf: All-pass RateArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of attempts where every atomic criterion passed · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	15.4% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
GDPval-AA v2.1 LeaderboardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Elo rating for performance on real-world work tasks · Anchored to DeepSeek V4.1 Flash (max) at 1600 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	Not reported	Not reported	Not reported	1,642.18 Elo Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
GPQA DiamondArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	91.21% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
Harvey LAB-AA v1.1: Hallucination-Gated All-Pass RateArtificial Analysis · Stirrup; Harvey private dataset v1.1.0; 120 tasks across 24 practice areas; up to 200 turns; internet disabled; GPT-6 Sol, Grok 4.7 and Claude Opus 5.5 rubric judges; GPT-6 Sol High two-pass hallucination check; hallucination-gated all-pass averaged across judges Benchmark details Average over tasks of the share of the judge panel finding every rubric criterion passed, with a task zeroed on any material hallucination · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	0.83% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	39.85% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
ITBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Combined public and private Site Reliability Engineering (SRE) tasks · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	51.24% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of tasks judged accurate, complete, and concise · Expert + compound (hard) sets · Benchmark developed by Wisedocs · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	51.11% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis Unit: index. Source collected 2026-10-18.	Not reported	Not reported	Not reported	7.47 Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader Benchmark details AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	27.5% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
SciCodeArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark curated by scientists across 16 disciplines. SciCode is under review following an independent audit of the dataset · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	51.62% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
𝜏³-BankingArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	47.22% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
Terminal-Bench-Science 0.1Artificial Analysis · mini-swe-agent; Terminal-Bench-Science 0.1.0; 70 tasks; pass@1 averaged over 3 repeats Benchmark details Share of tasks passed (pass@1), averaged over three repeats · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	4.76% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
Terminal-Bench v2.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	84.27% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
Terminal-Bench v4.0Artificial Analysis · mini-SWE-agent v2.4.6; 66 tasks; pass@1 averaged over 3 repeats; 500 steps; task timeout up to 8 hours Benchmark details Benchmark developed by the Laude Institute, Stanford, and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	32.83% Reported settings GLM-5.3-Flash; Reported effort: reasoning; Model source: https://artificialanalysis.ai/models/glm-5-3-flash; Mode from evaluator isReasoning metadata; effort budget not specified	Not reported	Not reported
ARC-AGI-2ARC Prize · v2 Semi-Private Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	27.92% Reported settings GLM-5.3-Flash (Low); Source model ID: zai-glm-5-3-flash-low; 120 tasks	50.14% Reported settings GLM-5.3-Flash (High); Source model ID: zai-glm-5-3-flash-high; 120 tasks	65.83% Reported settings GLM-5.3-Flash (Max); Source model ID: zai-glm-5-3-flash-max; 120 tasks	Not reported	Not reported	Not reported
DeepSWE v1.1Datacurve · mini-swe-agent Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	63.4% Reported settings glm-5-3-flash; Reported effort: max; 113 unique tasks; 448 attempts; 4 runs	Not reported	Not reported	Not reported
FrontierCode 1.1 · extended-chiselCognition · chisel Benchmark details FrontierCode 1.1; chisel. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	46.21% Reported settings FrontierCode 1.1; extended; 150 tasks; chisel; source effort max; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	Not reported	Not reported	Not reported
FrontierCode 1.1 · main-chiselCognition · chisel Benchmark details FrontierCode 1.1; chisel. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	31.83% Reported settings FrontierCode 1.1; main; 100 tasks; chisel; source effort max; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	Not reported	Not reported	Not reported
Terminal-Bench 4.0 · claudecodeTerminal-Bench · Claude Code Benchmark details Terminal-Bench 4.0; Claude Code. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	Not reported	35.8% Reported settings Terminal-Bench 4.0.0; 66 tasks; Claude Code; effort none; reported margin ±3.5 percentage points, confidence level unspecified; release date 2026-08-26; board checked 2026-10-07	Not reported

What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 8,915 results across 259 models · Collected 2026-10-18

GLM-5.3-Flash · Reasoning

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results