RLTasksby ulam.ai
Models / Z.ai

Z.ai / MODEL EVIDENCE

GLM-5.3

104 benchmarks · 213 observations · 91 sources

Scores at a glance

96 benchmarks · large dot: GLM-5.3; small dots: closest models where reported

AA-Briefcase v1.1 ↗1503.99 Elo
833.061811.82

GLM-5.3 (Max) · AA-Briefcase v1.1 · overall Elo ↗

1 more reported result

1507 Elo · GLM-5.3 (max) · AA-Briefcase v1.1 · AA-Briefcase v1.1 (Elo-score) · Kimi K3.1 Pro · complete release comparison · Introducing Kimi K3.1 Pro ↗

AutomationBench-AA ↗62.20%
4.9877.51

GLM-5.3 (Max) · AutomationBench-AA · guardrail-gated objective score ↗

1 more reported result

62.2% · GLM-5.3 (max) · AutomationBench-AA · Automation Bench-AA · Kimi K3.1 Pro · complete release comparison · Introducing Kimi K3.1 Pro ↗

GDPval-AA v2.1 ↗1649.45 Elo
1071.981866.27

GLM-5.3 (Max) · GDPval-AA v2.1 · Elo ↗

1 more reported result

1650 Elo · GLM-5.3 (max) · GDPval-AA v2.1 · GDPval-AA v2.1 (Elo-score) · Kimi K3.1 Pro · complete release comparison · Introducing Kimi K3.1 Pro ↗

AA-3D ↗783.01 Elo
572.551198.83

GLM-5.3 (Max) · AA-3D · Elo ↗

HealthBench ↗52.80%
50.765.98

GLM-5.3 (Max) · HealthBench · HealthBench Professional · length-adjusted ↗

AA-PersonalityBench ↗97.52 percentile
28.92100

GLM-5.3 (Max) · AA-PersonalityBench · expressiveness ↗

7 more reported results

20.80 percentile · GLM-5.3 (Max) · AA-PersonalityBench · organization · Artificial Analysis ↗

55.45 percentile · GLM-5.3 (Max) · AA-PersonalityBench · informativeness · Artificial Analysis ↗

80.20 percentile · GLM-5.3 (Max) · AA-PersonalityBench · transparency · Artificial Analysis ↗

90.10 percentile · GLM-5.3 (Max) · AA-PersonalityBench · warmth · Artificial Analysis ↗

90.10 percentile · GLM-5.3 (Max) · AA-PersonalityBench · conviction · Artificial Analysis ↗

92.58 percentile · GLM-5.3 (Max) · AA-PersonalityBench · imagination · Artificial Analysis ↗

30.70 percentile · GLM-5.3 (Max) · AA-PersonalityBench · slop · Artificial Analysis ↗

LMCA ↗55.51%
2.7868.26

GLM-5.3 (unknown thinking) · LMCA · Score · % ↗

DTBench ↗87.67%
41.6598.93

GLM-5.3 (unknown thinking) · DTBench · Accuracy · % ↗

SAFE-Teen ↗51%
1556

GLM 5.3 · Urgent safety · share of applicable ratings marked fail ↗

2 more reported results

70% · GLM 5.3 · Health advice · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗

64% · GLM 5.3 · Relationship boundaries · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗

SWE-Marathon ↗42.5%
4051.7

GLM-5.3 (max) · SWE-Marathon · SWE Marathon v1.1 · Kimi K3.1 Pro · complete release comparison ↗

PostTrainBench ↗34.5%
30.544.6

GLM-5.3 (max) · PostTrainBench · PostTrain Bench v1.2 · Kimi K3.1 Pro · complete release comparison ↗

Kimi Code Bench 2.0 ↗61.7%
61.777.1

GLM-5.3 (max) · Kimi Code Bench 2.0 · Kimi Code Bench 2.0 (Internal) · Kimi K3.1 Pro · complete release comparison ↗

Kimi Code Bench 3.0 ↗45.5%
45.566.5

GLM-5.3 (max) · Kimi Code Bench 3.0 · Kimi Code Bench 3.0 Long-horizon (Internal) · Kimi K3.1 Pro · complete release comparison ↗

BioMysteryBench ↗79.6%
78.981.9

GLM-5.3 (max) · BioMysteryBench · Kimi K3.1 Pro · complete release comparison ↗

HANDBOOK.md ↗20.8%
20.844.6

GLM-5.3 (max) · HANDBOOK.md · Kimi K3.1 Pro · complete release comparison ↗

DECK-Arena ↗1119 Elo
11191456

GLM-5.3 (max) · DECK-Arena · DECK-Arena (Internal, Elo-score) · Kimi K3.1 Pro · complete release comparison ↗

Design Bench ↗718 Elo
7181076

GLM-5.3 (max) · Design Bench · Design Bench (Internal, Elo-score) · Kimi K3.1 Pro · complete release comparison ↗

Kimi Hallucination Benchmark ↗22.6%
5.523.3

GLM-5.3 (max) · Kimi Hallucination Benchmark · Kimi Hallu Bench Hard w/ tools (Internal, lower is better) · Kimi K3.1 Pro · complete release comparison ↗

1 more reported result

14.8% · GLM-5.3 (max) · Kimi Hallucination Benchmark · Kimi Hallu Bench w/o tools (Internal, lower is better) · Kimi K3.1 Pro · complete release comparison · Introducing Kimi K3.1 Pro ↗

ExploitBench ↗54.4%
54.4100

GLM-5.3 (max) · ExploitBench · Kimi K3.1 Pro · complete release comparison ↗

ExploitGym ↗15%
1542.4

GLM-5.3 (max) · ExploitGym · Kimi K3.1 Pro · complete release comparison ↗

BTF-3 ↗0.149 Brier scale
0.120.15

GLM-5.3 · Pooled forecasting score ↗

GDPx v1.0 ↗2.7%
1.6712.03

GLM 5.3 · GDPx v1.0 · Overall · intended grading ↗

53 more reported results

2.8% · GLM 5.3 · GDPx v1.0 · Eight-domain subset · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.1% · GLM 5.3 · GDPx v1.0 · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.1% · GLM 5.3 · GDPx v1.0 · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.4% · GLM 5.3 · GDPx v1.0 · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

11.0% · GLM 5.3 · GDPx v1.0 · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

1.2% · GLM 5.3 · GDPx v1.0 · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.8% · GLM 5.3 · GDPx v1.0 · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.9% · GLM 5.3 · GDPx v1.0 · Overall · partial credit · GDPx v1.0 — SemiAnalysis ↗

8.9% · GLM 5.3 · GDPx v1.0 · Eight-domain subset · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.1% · GLM 5.3 · GDPx v1.0 · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.2% · GLM 5.3 · GDPx v1.0 · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.4% · GLM 5.3 · GDPx v1.0 · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.7% · GLM 5.3 · GDPx v1.0 · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

27.3% · GLM 5.3 · GDPx v1.0 · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

19.8% · GLM 5.3 · GDPx v1.0 · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

11.0% · GLM 5.3 · GDPx v1.0 · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

1.2% · GLM 5.3 · GDPx v1.0 · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

4.7% · GLM 5.3 · GDPx v1.0 · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Aria · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Arcverex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.2% · GLM 5.3 · GDPx v1.0 · Ergodex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.4% · GLM 5.3 · GDPx v1.0 · Arithmetic · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · LatchBio · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Living Context · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Chakra · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

11.0% · GLM 5.3 · GDPx v1.0 · Third Moment · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.1% · GLM 5.3 · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

1.2% · GLM 5.3 · GDPx v1.0 · Beautiful Machines · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · ValenceForge · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.6% · GLM 5.3 · GDPx v1.0 · Emulated · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Fleet · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Datacurve · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.5% · GLM 5.3 · GDPx v1.0 · Aria · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Arcverex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.2% · GLM 5.3 · GDPx v1.0 · Ergodex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.4% · GLM 5.3 · GDPx v1.0 · Arithmetic · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.7% · GLM 5.3 · GDPx v1.0 · LatchBio · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

27.3% · GLM 5.3 · GDPx v1.0 · Living Context · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

19.8% · GLM 5.3 · GDPx v1.0 · Chakra · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

11.0% · GLM 5.3 · GDPx v1.0 · Third Moment · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.1% · GLM 5.3 · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

1.2% · GLM 5.3 · GDPx v1.0 · Beautiful Machines · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.2% · GLM 5.3 · GDPx v1.0 · ValenceForge · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.2% · GLM 5.3 · GDPx v1.0 · Emulated · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Fleet · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · GLM 5.3 · GDPx v1.0 · Datacurve · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

19.2 percentile-point gap

53 of 66 shared benchmark comparisons

21.1 percentile-point gap

41 of 66 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.

POST-TRAINING / SOURCE DISCLOSURE

What the GLM-5.2 / GLM-5.3 report says

Z.ai says GLM-5.3 keeps the GLM-5.2 base and scales post-training on the same long-horizon stack: SAO, asynchronous slime training and a larger mix of environments and tasks.

  • Shared GLM-5.2 base checkpoint
  • Single-rollout asynchronous optimization (SAO)
  • Expanded long-horizon environment training
Single-rollout asynchronous optimization →Stateful agent training →

The launch report attributes the change to post-training, but its headline scores are publisher comparisons rather than a complete independent replication or an isolated optimizer ablation.

  1. GLM-5.3: Frontier Coding with Emergent Cyber Capabilities ↗

    Z.ai · 2026

Filter comparisons to this model →

Reports featuring this model

Kimi K3.1 Pro · Kimi →

DeepSeek V4.1 Flash 1017 · DeepSeek →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. Arena leaderboard — WebDev evaluation ↗

    Arena

  2. AA-3D: Interactive 3D Experience Benchmark | Artificial Analysis ↗

    Artificial Analysis

  3. AA-AnalystAgent Benchmark | Artificial Analysis ↗

    Artificial Analysis

  4. AA-Briefcase v1.1: Agentic Knowledge Work Benchmark | Artificial Analysis ↗

    Artificial Analysis

  5. AA-PersonalityBench: AI Personality & Communication Style Benchmark | Artificial Analysis ↗

    Artificial Analysis

  6. Artificial Analysis Intelligence Index v5.0 — Artificial Analysis ↗

    Artificial Analysis

  7. AA-LCR v1.1: Long Context Reasoning Benchmark | Artificial Analysis ↗

    Artificial Analysis

  8. AutomationBench-AA: Agentic SaaS Workflow Benchmark | Artificial Analysis ↗

    Artificial Analysis

  9. CritPt Benchmark | Artificial Analysis ↗

    Artificial Analysis

  10. CritPt v2 — Hard Subset — Artificial Analysis ↗

    Artificial Analysis

  11. EnterpriseOps-Gym-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  12. FrontierCode 1.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  13. FrontierCode 2.0 Benchmark — Artificial Analysis ↗

    Artificial Analysis

  14. GDP.pdf Benchmark | Artificial Analysis ↗

    Artificial Analysis

  15. GDPval-AA v2.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  16. GPQA Diamond Benchmark | Artificial Analysis ↗

    Artificial Analysis

  17. Harvey LAB-AA v1.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  18. HealthBench Professional Benchmark | Artificial Analysis ↗

    Artificial Analysis

  19. HLE-Diamond Benchmark — Artificial Analysis ↗

    Artificial Analysis

  20. Humanity's Last Exam Benchmark | Artificial Analysis ↗

    Artificial Analysis

  21. ITBench-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  22. MLCR-AA: Medical Long Context Reasoning Benchmark | Artificial Analysis ↗

    Artificial Analysis

  23. AA-Omniscience v1.1: Knowledge and Hallucination Benchmark — Artificial Analysis ↗

    Artificial Analysis

  24. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  25. 𝜏³-Banking Benchmark | Artificial Analysis ↗

    Artificial Analysis

  26. Terminal-Bench-Science 0.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  27. Terminal-Bench 2.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  28. Terminal-Bench 4.0 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  29. TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts — published result table ↗

    Jaffe, Oliver, Sherburn, Dane · 2026

  30. FrontierCode Leaderboard ↗

    Cognition

  31. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  32. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  33. Cursor · CursorBench ↗

    Cursor

  34. DeepSWE ↗

    Datacurve

  35. APEX-Agents ↗

    Epoch AI

  36. Chess Puzzles ↗

    Epoch AI

  37. CritPt v1 ↗

    Epoch AI

  38. FrontierMath Tier 4 (v2) ↗

    Epoch AI

  39. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  40. GPQA Diamond ↗

    Epoch AI

  41. Mystery Game Puzzles ↗

    Epoch AI

  42. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  43. SimpleQA Verified ↗

    Epoch AI

  44. Vending Bench 2 — methodology and published results ↗

    Epoch AI · 2026

  45. AI benchmarks & capabilities ↗

    Epoch AI

  46. FutureSearch Evals ↗

    FutureSearch

  47. GDPx v1.0: The General Agentic Benchmark — SemiAnalysis ↗

    SemiAnalysis · 2026

  48. WeirdML ↗

    Håvard Tveit Ihle

  49. deepseek-ai/DeepSeek-V4.1-Flash-1017 · Hugging Face ↗

    DeepSeek AI

  50. HiL-Bench (Human-in-Loop Benchmark) — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  51. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  52. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  53. SWE-Bench Pro V2 — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  54. SWE-Bench Pro V2 — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  55. FrontierSWE — software engineering benchmark ↗

    FrontierSWE

  56. Introducing Kimi K3.1 Pro ↗

    Moonshot AI · 2026

  57. Code Migration v1 — evaluation methodology and results ↗

    Vals AI · 2026

  58. Code Migration — evaluation methodology and results ↗

    Vals AI · 2026

  59. CyberBench — evaluation methodology and results ↗

    Vals AI · 2026

  60. Excel Modeling Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  61. Excel Modeling Benchmark — evaluation methodology and results ↗

    Vals AI · 2026

  62. Finance Agent v2 — evaluation methodology and results ↗

    Vals AI · 2026

  63. Finance Agent — evaluation methodology and results ↗

    Vals AI · 2026

  64. Vals Finance Index — evaluation methodology and results ↗

    Vals AI · 2026

  65. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  66. HLAB-Vals-1.1 — evaluation methodology and results ↗

    Vals AI · 2026

  67. Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  68. IOI — evaluation methodology and results ↗

    Vals AI · 2026

  69. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  70. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  71. Vals Legal Index — evaluation methodology and results ↗

    Vals AI · 2026

  72. Legal Research Bench — evaluation methodology and results ↗

    Vals AI · 2026

  73. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  74. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  75. SAFE-Teen: Teen Mental Health & Safety | Vals AI A conversation branches after the chatbot turns down a request — evaluation methodology and results ↗

    Vals AI · 2026

  76. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  77. MysteryMechanism — evaluation methodology and results ↗

    Vals AI · 2026

  78. ProgramBench — evaluation methodology and results ↗

    Vals AI · 2026

  79. ProofBench — evaluation methodology and results ↗

    Vals AI · 2026

  80. Public Benefits Bench — evaluation methodology and results ↗

    Vals AI · 2026

  81. Vals RSI Index — evaluation methodology and results ↗

    Vals AI · 2026

  82. SkillsBench — evaluation methodology and results ↗

    Vals AI · 2026

  83. SRE Bench — evaluation methodology and results ↗

    Vals AI · 2026

  84. SWE-bench Verified — evaluation methodology and results ↗

    Vals AI · 2026

  85. Tax Agent Bench — evaluation methodology and results ↗

    Vals AI · 2026

  86. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

  87. Terminal-Bench 4.0 — evaluation methodology and results ↗

    Vals AI · 2026

  88. Terminal-Bench-Science — evaluation methodology and results ↗

    Vals AI · 2026

  89. Vals Index — evaluation methodology and results ↗

    Vals AI · 2026

  90. Vibe Code Bench 1-100 — evaluation methodology and results ↗

    Vals AI · 2026

  91. Vibe Code Bench — evaluation methodology and results ↗

    Vals AI · 2026

Release dates reported by sources

2026-08-14 · 2026-08-18

Names used in source tables

GLM 5.3 · GLM 5.3 (Mini-SWE-agent) · GLM 5.3 (OpenHands) · GLM 5.3 (max) · GLM-5.3 · GLM-5.3 (Max) · GLM-5.3 (high) · GLM-5.3 (low) · GLM-5.3 (max) · GLM-5.3 (mini-swe-agent) max · GLM-5.3 (unknown thinking) · glm-5.3