RankingEffort benchmarks

GLM-5.3 · Max effort benchmarks

Compare measured Low, Medium, High and other settings. Every number keeps its source.

Model

GLM-5.3

38 measured results · 3 reported effort labels · 36 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

GLM-5.3 · Max: 57.5 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?

Cognition · 17.5% direct comparison weight · independent
Datacurve · 17.1% direct comparison weight · independent
Artificial Analysis · 17.0% direct comparison weight · independent
Epoch AI · 16.9% direct comparison weight · independent
Vals AI · 16.3% direct comparison weight · independent
Terminal-Bench · 15.2% direct comparison weight · independent

These shares describe the comparisons entering the fit, not fractions of the final score.

How sensitive is this score?

Remove one evaluator or family and refit the graph against the same overall panel. Values use the matching dataset audited 2026-09-12. These are sensitivity checks, not confidence intervals; ranking eligibility may also change.

Removed evidence	Recomputed score	Change
Artificial Analysis	58.7	+1.2
ARC Prize	59.0	+1.5
Datacurve	47.7	-9.8
LiveCodeBench	57.5	-0.0
Scale	58.5	+1.0
Terminal-Bench	57.1	-0.4
Cognition	55.7	-1.8
LiveBench	58.4	+0.9
SWE-rebench	57.1	-0.4
Epoch AI	63.2	+5.7
Vals AI	57.9	+0.4
Provider reports · Anthropic	57.4	-0.1
Provider reports · Moonshot AI	57.3	-0.2
Provider reports · Meta	57.5	+0.0
Provider reports · Mistral	57.5	+0.0
All provider reports	57.3	-0.3

Every benchmark family

Removed evidence	Recomputed score	Change
deepswe	47.8	-9.7
frontiercode	55.7	-1.8
arc-agi	59.0	+1.4
simpleqa	58.7	+1.2
mmlu-pro	58.7	+1.1
vals-legal-research	56.7	-0.9
mmmu-pro	58.4	+0.8
vals-excel-modeling	58.1	+0.5
harvey	57.0	-0.5
enterprise-ops	58.0	+0.5
gdp-pdf	58.0	+0.4
frontiermath	57.1	-0.4
swe-rebench	57.1	-0.4
gpqa	57.1	-0.4
tau-bench	57.2	-0.3
omniscience	57.9	+0.3
epoch-game-puzzles	57.2	-0.3
hle	57.8	+0.3
livebench-data	57.8	+0.3
vibe-code	57.2	-0.3
multi-swe-bench	57.2	-0.3
automationbench	57.2	-0.3
vals-finance-agent	57.2	-0.3
swe-atlas-qna	57.3	-0.2
ifbench	57.3	-0.2
arc-agi-3	57.3	-0.2
briefcase	57.3	-0.2
vals-code-migration	57.4	-0.2
gdpval	57.4	-0.1
swe-atlas-test-writing	57.4	-0.1
livebench-instructions	57.7	+0.1
scicode	57.4	-0.1
livebench-coding	57.4	-0.1
aa-lcr	57.4	-0.1
livebench-reasoning	57.4	-0.1
livebench-language	57.6	+0.1
terminal-bench	57.4	-0.1
critpt	57.6	+0.1
enigma-eval	57.6	+0.1
aime	57.6	+0.0
itbench	57.5	-0.0
apex-agents	57.6	+0.0
swe-atlas-refactoring	57.5	-0.0
analyst-agent	57.5	-0.0
livebench-math	57.5	-0.0
livecodebench	57.5	+0.0
gmmlu	57.5	+0.0
math500	57.5	+0.0
swe-bench-pro	57.5	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 60 observations, 34 contributing, 0 matched without graph weight and 26 excluded. Counts do not establish rank eligibility.

1 · Benchmark family or source protocol has not been reviewed
3 · Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode.
7 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
1 · No matched, identified effort unit; duplicates and unnamed comparators do not add evidence
13 · Supporting evidence outside the reviewed capability core
1 · The GLM card attributes this evaluation to Artificial Analysis; it is not a new Z.ai comparison.

Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AGLM-5.3 · Max4 of 7 capabilities supported

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

Agentic
A60.1Supported
Hard reasoning
A43.0Supported
Coding
A67.6Supported
Human pref
Unknown
Knowledge
A23.7Supported
Multimodal
Unknown
Long context
A70.2Preliminary

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.

Compare this configuration →Explore the model evidence profile →

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
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase 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 2026-09-12.	Not reported	Not reported	1,512.51 Elo Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	79.67% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	62.2% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	19.14% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	36.44% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
GDP.pdf: All-passArtificial 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 2026-09-12.	Not reported	Not reported	11.2% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
GDPval-AA v2 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 a human baseline of 1,000 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-09-12.	Not reported	Not reported	1,667.27 Elo Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	91.72% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	42.26% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	48.33% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	14.3 Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	33.85% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
SciCodeArtificial 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 2026-09-12.	Not reported	Not reported	59.03% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	50.31% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	83.9% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
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 2026-09-12.	Not reported	Not reported	41.92% Reported settings GLM-5.3 (max); Model source: https://artificialanalysis.ai/models/glm-5-3
DeepSWE v1.1Datacurve · mini-swe-agent Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	69% Reported settings glm-5-3; Reported effort: max; 113 unique tasks; 451 attempts; 4 runs
VulcanBench v3 · Report 18VulcanBench · VulcanBench bare-bones API loop, deterministic hidden tests in Docker Benchmark details Frozen v3, 23 post-cutoff repository tasks. Published pass@1; repeat counts and time budgets vary by report. Display only pending comparability review. Unit: percent. Source collected 2026-09-12.	78.3% Reported settings GLM 5.3; source label: low; low effort; 23 tasks; evaluation 2026-08-24. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source label needs review	73.9% Reported settings GLM 5.3; source label: high; high effort; 23 tasks; evaluation 2026-08-24. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source label needs review	65.2% Reported settings GLM 5.3; source label: max; max effort; 23 tasks; evaluation 2026-08-24. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source labels low/high/max; Z.ai API effort labels may be metadata only. Bare-bones API column, not ZCode. Source label needs review
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 2026-09-12.	Not reported	Not reported	52.63% 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
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 2026-09-12.	Not reported	Not reported	40.14% 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
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 2026-09-12.	Not reported	Not reported	41.8% Reported settings Terminal-Bench 4.0.0; 66 tasks; Claude Code; effort max; reported margin ±3.2 percentage points, confidence level unspecified; release date 2026-08-14; board checked 2026-09-07
Epoch · Chess Puzzles · v1.1.6Epoch AI · Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	21% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=glm-5.3_max; run=9RfS3yHmzmP3azirghceFL; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · FrontierMath Tier 4 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	29.27% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=glm-5.3_max; run=4S3ZrJDkwh4gmGrmJQdvzR; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · FrontierMath Tiers 1–3 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	68.77% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=glm-5.3_max; run=iXvqUzwxAiA3Rg3dwnutFs; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · GPQA diamond · v1.0.11Epoch AI · Epoch AI Inspect; task 1.0.11; scorer choice Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	90.91% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=glm-5.3_max; run=dLDBahYBTGRNjbsXA5pyr3; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · Mystery Game Puzzles · v1.0.4 · Tool submissionEpoch AI · Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	33% Reported settings Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer; model=glm-5.3_max; run=UUq99Qhq3qizcoJo5vdhuK; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · OTIS Mock AIME 2024-2025 · v1.1.6Epoch AI · Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	91.11% Reported settings Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=glm-5.3_max; run=goZR5yGxZic9TZ23ezCL3w; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · SimpleQA Verified · v1.2.0Epoch AI · Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	41% Reported settings Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer; model=glm-5.3_max; run=TY8R8FV7zCwByEipkqyg2P; mean accuracy; https://epoch.ai/data/benchmarks.csv
Vals AI · Finance Agent v2Vals AI · Vals FAB v2 shared six-tool harness; two-hour task limit Benchmark details Overall weighted rubric accuracy, averaged over three runs on 450 tasks. Dealbreaker failures and timeouts score zero. Related categories and all-pass scores are not additional votes. Three-model judge panel; private test set. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	55.84% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":131072,"verbosity":null,"provider":"Fireworks AI","harness":null}
Vals AI · Code Migration v1Vals AI · Vals Code Migration shared agent harness Benchmark details Hidden behavior-test pass rate, equally weighted across 40 repositories (30 CLI and 10 COBOL-to-Java). Anti-cheat failures score zero. Code-quality ratings and language splits are not additional votes. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	44.22% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}
Vals AI · Legal Research Bench v1Vals AI · Vals Legal Research shared five-tool harness; three-hour task limit Benchmark details All-pass accuracy on 208 private legal research tasks. Every expert rubric check must pass; GPT-5.4 judges responses. Partial-credit scores and practice-area splits are not additional votes. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	49.04% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}
Vals AI · Excel Modeling Benchmark v1Vals AI · Vals EMB shared workbook agent harness; 3.5-hour task limit Benchmark details Mean of seven equally weighted task categories. Template submissions use recalculated Excel cell accuracy; scratch submissions use rubric checks judged by GPT-5.4 Mini XHigh. Partial credit does not mean task completion. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	56.34% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}
Vals AI · MMLU-Pro v1Vals AI · Vals MMLU-Pro five-shot chain-of-thought; regex answer grading Benchmark details Mean accuracy across 14 subjects under Vals five-shot prompting. Subject splits share one family and are not additional votes. Vals labels this benchmark archived; it is historical corroboration, not ongoing frontier evaluation. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	86.77% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-01","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}
Vals AI · Terminal-Bench 2.1Vals AI · Vals Terminus 2 harness; pass@1 on 89 tasks Benchmark details Overall task pass rate under the shared Terminus 2 configuration. All task tests must pass. Difficulty splits share one family. Fable 5 and Opus 5 mixed-checkpoint fallback runs are excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	71.54% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}
Vals AI · Vibe Code Bench v1.1Vals AI · Vals modified OpenHands harness; five hours or 1,000 turns per app Benchmark details Mean app test pass rate on the canonical v1.1 test set. Browser Use executes workflow tests; each test requires at least 90% of its substeps. Automated judging can introduce error. Alternate agents are excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	78.13% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":131072,"verbosity":null,"provider":"Fireworks AI","harness":"OpenHands"}
Vals AI · Harvey Legal Agent Benchmark v1Vals AI · Vals Harvey held-out legal file-work protocol; internet disabled Benchmark details Mean task resolution across two judges (GPT-5.5 Medium and Sonnet 4.6), requiring every criterion to pass. Criteria pass rates and practice-area splits do not add votes. Shared Harvey family; Fable 5 fallback run excluded. Unit: percent. Source collected 2026-09-12.	Not reported	Not reported	8.33% Reported settings {"checkpoint":"glm-5.3","evaluatorModel":"zai/glm-5.3","sourceUpdatedOn":"2026-09-10","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":0.95,"max_output_tokens":128000,"verbosity":null,"provider":"Zhipu AI","harness":null}

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 · 7,424 results across 240 models · Collected 2026-09-12

GLM-5.3 · Max effort benchmarks

GLM-5.3

Capability score evidence

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context