GLM-5.3-Flash · Intelligence Index v4.3.2 ↗
Z.ai / MODEL EVIDENCE
GLM-5.3-Flash
62 benchmarks · 75 observations · 66 sources
Scores at a glance
62 benchmarks · large dot: GLM-5.3-Flash; small dots: closest models where reported
GDPval-AA v2.1 ↗1642.16 Elo
GLM-5.3-Flash · GDPval-AA v2.1 · Elo ↗
AA-3D ↗936.61 Elo
GLM-5.3-Flash · AA-3D · Elo ↗
AA-AnalystAgent ↗25.00%
GLM-5.3-Flash · AA-AnalystAgent · strict score ↗
Harvey LAB-AA v1.1 ↗0.83%
GLM-5.3-Flash · Harvey LAB-AA v1.1 · all-pass ↗
EnterpriseOps-Gym-AA ↗33.21%
GLM-5.3-Flash · EnterpriseOps-Gym-AA ↗
FrontierCode 1.1 ↗41.00%
GLM-5.3-Flash · FrontierCode 1.1 · AA implementation ↗
1 more reported result
31.83% · GLM-5.3-Flash (max) · FrontierCode · Main score · % · Epoch AI · FrontierCode ↗
AA-LCR v1.1 ↗80.00%
GLM-5.3-Flash · AA-LCR v1.1 ↗
AA-Omniscience v1.1 ↗7.47 index
GLM-5.3-Flash · AA-Omniscience · error-penalized score ↗
GLM-5.3-Flash · AA-Omniscience · accuracy ↗
GLM-5.3-Flash · AA-Omniscience · hallucination rate, lower is better ↗
SciCode ↗51.62%
GLM-5.3-Flash · SciCode · AA implementation ↗
Humanity’s Last Exam ↗39.85%
GLM-5.3-Flash · Humanity’s Last Exam · AA implementation ↗
GPQA Diamond ↗86.36 %
GLM 5.3 Flash · GPQA Diamond · Accuracy ↗
2 more reported results
91.21% · GLM-5.3-Flash · GPQA Diamond · AA implementation · Artificial Analysis ↗
90.15% · GLM-5.3-Flash (max) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
ITBench-AA ↗51.24%
GLM-5.3-Flash · ITBench-AA · root-cause score ↗
MLCR-AA ↗11.67%
GLM-5.3-Flash · MLCR-AA · overall ↗
τ³-Banking ↗47.22%
GLM-5.3-Flash · τ³-Banking ↗
Terminal-Bench 2.1 ↗84.27%
GLM-5.3-Flash · Terminal-Bench 2.1 ↗
CritPt ↗15.43%
GLM-5.3-Flash · CritPt · original ↗
ErdősBench ↗2.950 / 4
GLM-5.3 Flash · ErdősBench · mean progress / 4 · API ↗
GLM-5.3-Flash (max) · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
GLM-5.3-Flash (max) · FrontierMath-Tier-4-v2-Private · Best score (across scorers) · % ↗
GLM-5.3-Flash (max) · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
GLM-5.3-Flash (max) · Chess Puzzles · Best score (across scorers) · % ↗
GLM-5.3-Flash (max) · Mystery Game Puzzles · Best score (across scorers) · % ↗
ARC-AGI-1 ↗91%
GLM-5.3-Flash (max) · ARC-AGI · Score · % ↗
2 more reported results
71.83% · GLM-5.3-Flash (high) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
47% · GLM-5.3-Flash (low) · ARC-AGI · Score · % · Epoch AI · ARC-AGI ↗
ARC-AGI-2 ↗65.83%
GLM-5.3-Flash (max) · ARC-AGI-2 · Score · % ↗
2 more reported results
50.14% · GLM-5.3-Flash (high) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
27.92% · GLM-5.3-Flash (low) · ARC-AGI-2 · Score · % · Epoch AI · ARC-AGI-2 ↗
CursorBench 4.0 ↗36.8%
GLM-5.3-Flash (max) · CursorBench · Score · % ↗
2 more reported results
31.1% · GLM-5.3-Flash (high) · CursorBench · Score · % · Epoch AI · CursorBench ↗
26.9% · GLM-5.3-Flash (low) · CursorBench · Score · % · Epoch AI · CursorBench ↗
GLM-5.3-Flash (max) · Surface Evolver Bench · Mean score · % ↗
DeepSWE v1.1 ↗63.39%
GLM-5.3-Flash (max) · DeepSWE · Pass@1 · % ↗
ProofBench ↗21 %
GLM 5.3 Flash · ProofBench · Accuracy ↗
1 more reported result
21% · GLM-5.3-Flash (max) · ProofBench · Accuracy · % · Epoch AI · ProofBench ↗
GDP.pdf ↗14%
GLM-5.3-Flash (max) · GDP.pdf · GDP.pdf score · % ↗
FrontierSWE v2 ↗18.13%
GLM-5.3-Flash (max) · FrontierSWE · Score · % ↗
Epoch Capabilities Index (ECI) ↗153.08 ECI
GLM-5.3-Flash · General ECI · 26 October 2026 fit ↗
Vals Index ↗50.86 %
GLM 5.3 Flash · Vals Index · Accuracy ↗
Vals Finance Index ↗55.83 %
GLM 5.3 Flash · Vals Finance Index · Accuracy ↗
Vals Legal Index ↗71.05 %
GLM 5.3 Flash · Vals Legal Index · Accuracy ↗
Legal Research Bench ↗45.19 %
GLM 5.3 Flash · Legal Research Bench · All-Pass ↗
HLAB-Vals-1.1 ↗10.42 %
GLM 5.3 Flash · HLAB-Vals-1.1 · Task Pass Rate ↗
LegalBench ↗83.93 %
GLM 5.3 Flash · LegalBench · Accuracy ↗
GLM 5.3 Flash · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Vals Finance Agent v2.1 ↗60.59 %
GLM 5.3 Flash · Finance Agent · Accuracy ↗
GLM 5.3 Flash · Excel Modeling Benchmark · Accuracy ↗
1 more reported result
55.93 % · GLM 5.3 Flash · Excel Modeling Benchmark v1 · Accuracy · Vals AI · Excel Modeling Benchmark v1 ↗
Tax Agent Bench ↗31.55 %
GLM 5.3 Flash · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗57.85 %
GLM 5.3 Flash · Finance Agent v2 · Accuracy ↗
MortgageTax & TaxEval v2 ↗75.59 %
GLM 5.3 Flash · TaxEval · Accuracy ↗
MedScribe ↗88.94 %
GLM 5.3 Flash · MedScribe · Accuracy ↗
GLM 5.3 Flash (Mini-SWE-agent) · Terminal-Bench-Science · Score ↗
MMLU-Pro ↗86.06 %
GLM 5.3 Flash · MMLU Pro · Accuracy ↗
MMMU-Pro ↗86.01 %
GLM 5.3 Flash · MMMU Pro · Accuracy ↗
Vibe Code Bench ↗16.03 %
GLM 5.3 Flash · Vibe Code Bench · Vibe Code Bench 1–100 ↗
1 more reported result
30.76 % · GLM 5.3 Flash (OpenHands) · Vibe Code Bench · original task set · Vals AI · Vibe Code Bench ↗
Code Migration v1.1 ↗51.65 %
GLM 5.3 Flash · Code Migration · Accuracy ↗
1 more reported result
20.52 % · GLM 5.3 Flash · Code Migration v1 · Accuracy · Vals AI · Code Migration v1 ↗
GLM 5.3 Flash · ProgramBench · Accuracy ↗
Terminal-Bench 4.0 ↗37.37 %
GLM 5.3 Flash (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
SkillsBench ↗40.16 %
GLM 5.3 Flash (OpenHands) · SkillsBench · Score ↗
IOI v1 & IOI v2 ↗52.5 %
GLM 5.3 Flash · IOI · Accuracy ↗
GLM 5.3 Flash (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗80.51 %
GLM 5.3 Flash · LiveCodeBench · Accuracy ↗
CyberBench v1.1 ↗67.86 %
GLM 5.3 Flash · CyberBench · Accuracy ↗
SRE-Bench ↗0.76 %
GLM 5.3 Flash · SRE Bench · Accuracy ↗
GLM-5.3-Flash (max) · Humanity's Sixth Sense · Score ↗
BTF-3 ↗0.149 Brier scale
GLM-5.3-Flash · Pooled forecasting score ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
Kimi K3 →Kimi
35 of 51 shared benchmark comparisons
Grok 4.7 →xAI
37 of 51 shared benchmark comparisons
44 of 51 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- ARC Prize - Leaderboard ↗
- AA-3D: Interactive 3D Experience Benchmark | Artificial Analysis ↗
- AA-AnalystAgent Benchmark | Artificial Analysis ↗
- Artificial Analysis Intelligence Index v5.0 — Artificial Analysis ↗
- AA-LCR v1.1: Long Context Reasoning Benchmark | Artificial Analysis ↗
- CritPt Benchmark | Artificial Analysis ↗
- EnterpriseOps-Gym-AA Benchmark | Artificial Analysis ↗
- FrontierCode 1.1 Benchmark | Artificial Analysis ↗
- GDPval-AA v2.1 Benchmark | Artificial Analysis ↗
- GPQA Diamond Benchmark | Artificial Analysis ↗
- Harvey LAB-AA v1.1 Benchmark | Artificial Analysis ↗
- Humanity's Last Exam Benchmark | Artificial Analysis ↗
- ITBench-AA Benchmark | Artificial Analysis ↗
- MLCR-AA: Medical Long Context Reasoning Benchmark | Artificial Analysis ↗
- AA-Omniscience v1.1: Knowledge and Hallucination Benchmark — Artificial Analysis ↗
- SciCode Benchmark | Artificial Analysis ↗
- 𝜏³-Banking Benchmark | Artificial Analysis ↗
- Terminal-Bench 2.1 Benchmark | Artificial Analysis ↗
- FrontierCode Leaderboard ↗
- Cursor · CursorBench ↗
- DeepSWE ↗
- ARC-AGI-2 ↗
- Chess Puzzles ↗
- FrontierMath Tier 4 (v2) ↗
- FrontierMath Tiers 1-3 (v2) ↗
- GPQA Diamond ↗
- Mystery Game Puzzles ↗
- OTIS Mock AIME 2024-2025 ↗
- AI benchmarks & capabilities ↗
- FutureSearch Evals ↗
- Humanity's Sixth Sense — evaluation methodology and results ↗
- GDP.pdf Benchmark | Surge AI ↗
- FrontierSWE — software engineering benchmark ↗
- ErdősBench: research mathematics, scoped progress and proof audits ↗
- Code Migration v1 — evaluation methodology and results ↗
- Code Migration — evaluation methodology and results ↗
- CyberBench — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Excel Modeling Benchmark — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- Finance Agent — evaluation methodology and results ↗
- Vals Finance Index — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- HLAB-Vals-1.1 — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- IOI — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Vals Legal Index — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- MMMU Pro — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- SkillsBench — evaluation methodology and results ↗
- SRE Bench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Terminal-Bench 4.0 — evaluation methodology and results ↗
- Terminal-Bench-Science — evaluation methodology and results ↗
- Vals Index — evaluation methodology and results ↗
- Vibe Code Bench 1-100 — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
- Surface Evolver Bench ↗
Release dates reported by sources
2026-08-20 · 2026-08-26
Names used in source tables
GLM 5.3 Flash · GLM 5.3 Flash (Mini-SWE-agent) · GLM 5.3 Flash (OpenHands) · GLM-5.3 Flash · GLM-5.3-Flash · GLM-5.3-Flash (high) · GLM-5.3-Flash (low) · GLM-5.3-Flash (max)