GPQA Diamond ↗80.05 %
GLM 4.7 · GPQA Diamond · Accuracy ↗
1 more reported result
83.33% · GLM-4.7 · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗
Z.ai / MODEL EVIDENCE
GLM-4.7
26 benchmarks · 28 observations · 27 sources · released 2025-12-22
Scores at a glance
26 benchmarks · large dot: GLM-4.7; small dots: closest models where reported
GLM-4.7 · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
SimpleQA Verified ↗32.2%
GLM-4.7 · SimpleQA Verified · Best score (across scorers) · % ↗
GLM-4.7 · Chess Puzzles · Best score (across scorers) · % ↗
SimpleBench ↗47.7%
GLM-4.7 · SimpleBench · Score (AVG@5) · % ↗
Terminal-Bench 2.0 ↗33.39%
GLM-4.7 · Terminal Bench · Accuracy mean · % ↗
1 more reported result
33.26% · GLM-4.7 · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗
CritPt ↗1.71%
GLM-4.7 · CritPt · Accuracy · % ↗
GLM-4.7 · ProofBench · Accuracy · % ↗
SciCode ↗45.14%
GLM-4.7 · SciCode · Score · % ↗
Text Arena (Coding) ↗1434.7 Elo
GLM-4.7 · WebDev Arena · Elo ↗
Epoch Capabilities Index (ECI) ↗143.59 ECI
GLM-4.7 · General ECI · 26 October 2026 fit ↗
LegalBench ↗83.36 %
GLM 4.7 · LegalBench · Accuracy ↗
CaseLaw v2 ↗54.88 %
GLM 4.7 · CaseLaw · Accuracy ↗
CorpFin v2 ↗46.39 %
GLM 4.7 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗68.77 %
GLM 4.7 · TaxEval · Accuracy ↗
MedCode ↗32.77 %
GLM 4.7 · MedCode · Accuracy ↗
MedScribe ↗68.63 %
GLM 4.7 · MedScribe · Accuracy ↗
MedQA (USMLE) ↗93.74 %
GLM 4.7 · MedQA · Accuracy ↗
AIME 2024–2025 · Vals ↗93.33 %
GLM 4.7 · AIME · Accuracy ↗
MGSM (Multilingual GSM) ↗88.18 %
GLM 4.7 · MGSM · Accuracy ↗
MMLU-Pro ↗82.74 %
GLM 4.7 · MMLU Pro · Accuracy ↗
SWE-bench Verified ↗69.4 %
GLM 4.7 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗82.23 %
GLM 4.7 · LiveCodeBench · Accuracy ↗
NYU CTF Bench ↗19.8%
GLM-4.7 · NYU CTF Bench · Cotool 81-challenge subset · accuracy ↗
ALE-Bench ↗399.48 Performance points
glm-4.7 · Average contest performance ↗
Vending-Bench 2 ↗2376.816 USD
GLM-4.7 · End-of-year balance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
Qwen 3.5 Flash →Qwen
10 of 16 shared benchmark comparisons
DeepSeek V3.2 →DeepSeek
11 of 16 shared benchmark comparisons
10 of 16 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 ↗
- SciCode Benchmark | Artificial Analysis ↗
- Ale Bench — methodology and published results ↗
- Chess Puzzles ↗
- CritPt v1 ↗
- GPQA Diamond ↗
- OTIS Mock AIME 2024-2025 ↗
- SimpleQA Verified ↗
- SimpleBench ↗
- Vending Bench 2 — methodology and published results ↗
- AI benchmarks & capabilities ↗
- NYU CTF Bench — Cotool evaluation, February 2026 ↗
- 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 ↗
- MedCode — evaluation methodology and results ↗
- MedQA — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MGSM — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
Names used in source tables
GLM 4.7 · GLM 4.7 (Mini-SWE-agent) · GLM-4.7 · glm-4.7