GPQA Diamond ↗83.33 %
GLM 5 · GPQA Diamond · Accuracy ↗
1 more reported result
87.82% · GLM-5 · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
Z.ai / MODEL EVIDENCE
GLM-5
28 benchmarks · 30 observations · 30 sources · released 2026-02-11
Scores at a glance
28 benchmarks · large dot: GLM-5; small dots: closest models where reported
GLM-5 · FrontierMath-2025-02-28-Private · Best score (across scorers) · % ↗
GLM-5 · FrontierMath-Tier-4-2025-07-01-Private · Best score (across scorers) · % ↗
GLM-5 · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
SWE-bench Verified ↗72.08%
GLM-5 · SWE-Bench verified · Best score (across scorers) · % ↗
1 more reported result
71.4 % · GLM 5 (Mini-SWE-agent) · SWE-bench Verified · Score · Vals AI · SWE-bench Verified ↗
ARC-AGI-1 ↗44.67%
GLM-5 · ARC-AGI · Score · % ↗
SimpleBench ↗53.2%
GLM-5 · SimpleBench · Score (AVG@5) · % ↗
Terminal-Bench 2.0 ↗52.42%
GLM-5 · Terminal Bench · Accuracy mean · % ↗
WeirdML (v2) ↗48.17%
GLM-5 · WeirdML · Accuracy · % ↗
ARC-AGI-2 ↗4.86%
GLM-5 · ARC-AGI-2 · Score · % ↗
CL-bench ↗18.7%
GLM-5 · CL-bench · Overall · % ↗
Text Arena (Coding) ↗1435.91 Elo
GLM-5 · WebDev Arena · Elo ↗
Epoch Capabilities Index (ECI) ↗145.80 ECI
GLM-5 · General ECI · 26 October 2026 fit ↗
LegalBench ↗84.06 %
GLM 5 · LegalBench · Accuracy ↗
CaseLaw v2 ↗52.52 %
GLM 5 · CaseLaw · Accuracy ↗
CorpFin v2 ↗62.9 %
GLM 5 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗70.03 %
GLM 5 · TaxEval · Accuracy ↗
MedQA (USMLE) ↗94.27 %
GLM 5 · MedQA · Accuracy ↗
AIME 2024–2025 · Vals ↗91.67 %
GLM 5 · AIME · Accuracy ↗
MMLU-Pro ↗86.03 %
GLM 5 · MMLU Pro · Accuracy ↗
Vibe Code Bench ↗23.36 %
GLM 5 (OpenHands) · Vibe Code Bench · original task set ↗
LiveCodeBench ↗81.87 %
GLM 5 · LiveCodeBench · Accuracy ↗
DistressBench ↗71.17%
glm-5 · DistressBench · Score ↗
Glm-5 (Mini-SWE-Agent) · SWE Atlas - Refactoring · Score ↗
Glm 5 (Mini-SWE-Agent) · Scale Labs · Score ↗
ALE-Bench ↗765.62 Performance points
glm-5 · Average contest performance ↗
ForecastBench ↗63.7 Brier index
glm-5† · Overall difficulty-adjusted Brier index ↗
Vending-Bench 2 ↗4432.122 USD
GLM-5 · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
o3 →OpenAI
12 of 17 shared benchmark comparisons
Claude Sonnet 4.5 →Anthropic
13 of 17 shared benchmark comparisons
Grok 4 →xAI
12 of 17 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.
- Arena leaderboard — WebDev evaluation ↗
- Ale Bench — methodology and published results ↗
- ARC-AGI-1 ↗
- ARC-AGI-2 ↗
- CL-bench ↗
- Forecastbench — methodology and published results ↗
- FrontierMath Tier 4 (v1, old) ↗
- FrontierMath Tiers 1-3 (v1, old) ↗
- GPQA Diamond ↗
- OTIS Mock AIME 2024-2025 ↗
- SimpleBench ↗
- SWE-bench Verified ↗
- Vending Bench 2 — methodology and published results ↗
- WeirdML (v2) ↗
- AI benchmarks & capabilities ↗
- DistressBench — evaluation methodology and results ↗
- Scale Labs — evaluation methodology and results ↗
- SWE Atlas - Refactoring — evaluation methodology and results ↗
- Terminal-Bench 2.0 leaderboard ↗
- AIME — evaluation methodology and results ↗
- CaseLaw — evaluation methodology and results ↗
- CorpFin — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- MedQA — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
GLM 5 · GLM 5 (Mini-SWE-agent) · GLM 5 (OpenHands) · GLM-5 · Glm 5 (Mini-SWE-Agent) · Glm-5 (Mini-SWE-Agent) · glm-5 · glm-5†