Provider logo

GLM 4.7 Thinking

zai-org/glm-4.7:thinking
Provider logo

GLM 4.7 Thinking

zai-org/glm-4.7:thinking

GLM-4.7 with extended thinking capabilities for enhanced reasoning on complex tasks.

Added Dec 22, 2025

Model weights

Context Window

200.0K

Max Output

65.5K

Avg output tokens (7d)

1.7K tokens

65%

Input Price (Auto)

$0.42/1M

Output Price (Auto)

$1.54/1M

Cache Read (Auto)

$0.21/1M

Capabilities

Benchmarks

Performance metrics and benchmarks

Sourced from Artificial Analysis.

Intelligence Index

34.5

Better than 76% of models compared

Coding Index

45.3

Better than 58% of models compared

Agentic Index

26.2

Better than 58% of models compared

Reasoning

GPQA Diamond

Graduate-level scientific reasoning

85.9%

Better than 83% of models compared

HLE

Humanity's Last Exam

27.4%

Better than 82% of models compared

IFBench

Instruction-following benchmark

67.9%

Better than 81% of models compared

T²-Bench Telecom

Conversational AI agents in dual-control scenarios

95.9%

Better than 96% of models compared

AA-LCR

Long context reasoning evaluation

68.0%

Better than 75% of models compared

GDPval-AA

Economically valuable tasks

33.5%

CritPt

Research-level physics reasoning

1.7%

Coding

SciCode

Python programming for scientific computing

45.1%

Better than 83% of models compared

Terminal-Bench Hard

Agentic coding and terminal use

31.8%

Better than 75% of models compared

LiveCodeBench

Contamination-free coding benchmark

89.4%

Better than 99% of models compared

Math

AIME 2025

American Invitational Mathematics Examination 2025

95.0%

Better than 97% of models compared

Knowledge

MMLU-Pro

Professional and academic subject knowledge

85.6%

Better than 92% of models compared

AA-Omniscience Accuracy

Proportion of correctly answered questions

29.3%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

93.0%

Last updated Aug 16, 2026

Artificial Analysis

Providers

Choose explicit providers for this model. Auto routing remains available as the default option.

Loading provider options…

Compare GLM 4.7 Thinking with similar models from the same provider or model family.