supermarket

Language

GLM-4.7-Flash

By Z.ai (Zhipu AI), in the GLM family.

Released
19 Jan 2026
Status
Legacy
Context
204.8K
Max output
131.1K
Input per 1M tokens
$0.00
Output per 1M tokens
$0.00
Open weights
Yes
License
mit
Parameters
30B, 3B active
Knowledge cutoff
—
Reasoning
Yes
Modalities
text → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Knowledge & reasoning0.0Not ranked1 of 4
Human preference45.0Not ranked1 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Chess PuzzlesPuzzles solved0.0%Independentnone28 Aug 2026Epoch AI
Chess PuzzlesPuzzles solved0.0%Independent—28 Aug 2026Epoch AI
Humanity's Last Exam (HLE)Accuracy14.4%Self-reported——Model Card via Hugging Face Community Evals

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy60.5%Independent—30 Aug 2026Epoch AI
GPQA DiamondAccuracy45.1%Independentnone30 Aug 2026Epoch AI
GPQA DiamondAccuracy75.2%Self-reported——Model Card via Hugging Face Community Evals

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
OTIS Mock AIME 2024-2025Accuracy58.3%Independent—30 Aug 2026Epoch AI
OTIS Mock AIME 2024-2025Accuracy25.0%Independentnone28 Aug 2026Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Text Arena (arena.ai)Arena score (Elo scale)1365Independent—25 Sep 2026Arena
Text Arena (arena.ai)Votes11983Independent—25 Sep 2026Arena

Other GLM versions

Sources

GLM-4.7-Flash: benchmarks, price and specs · supermarket.dev