ModelBeat
Models/GLM-4.7
All models
Z

Z.ai (Zhipu AI): GLM-4.7

Watch this modelCompareAnnouncement
z-ai/glm-4.7

GLM-4.7 is Z.ai (Zhipu AI)’s open-weights AI model, released Dec 22, 2025. It costs $0.60/M input and $2.20/M output tokens, and handles a 205K-token context window. Its strongest use case is coding, where it ranks in the 60th percentile of the models Model Beat tracks.

GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.

Context
205K
Input
$0.60 / 1M
Output
$2.20 / 1M
Released
Dec 22, 2025
Modalities
text→text
Strongest atCoding top 40%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where GLM-4.7 lands among all models tracked on Model Beat.

39.0
Intelligence Index
Epoch AI
39th percentile of tracked models
60.0
Coding Index
Epoch AI
60th percentile of tracked models
42.0
Agentic Index
Epoch AI
42nd percentile of tracked models

Reasoning

5 evals
GPQA Diamond83.3%

Graduate-level scientific reasoning

Humanity's Last Exam25.1%

Frontier of human expert knowledge

MMLU-Pro85.6%

Broad expert knowledge

SimpleBench47.7%

Common-sense trick questions

SimpleQA Verified31.5%

Factual accuracy & hallucination

Coding

3 evals
LiveCodeBench89.4%

Contamination-free coding

SciCode45.1%

Scientific research coding

WebDev Arena1441

Human-rated web development

Math

3 evals
AIME 2024/202583.3%

Olympiad-qualifier math

FrontierMath2.4%

Research-level math problems

FrontierMath Tier 40.0%

Hardest research math

Agentic & Tools

3 evals
APEX8.7%

Multi-step agentic tasks

Terminal-Bench33.4%

Command-line agentic tasks

τ²-bench95.9%

Tool-agent-user reliability

In the news

7

Changelog

8
  • GLM-4.7: APEX improved from 3.1% to 8.7% (+181%)benchmark · Jul 30, 2026
  • GLM-4.7: cheapest credible provider via Google now $0.6 per 1M input tokens (was $0.52, +15%)cheapest provider · Jul 29, 2026
  • GLM-4.7: cheapest credible provider via Google now $2.2 per 1M output tokens (was $1.85, +19%)cheapest provider · Jul 29, 2026
  • GLM-4.7: cheapest credible provider via AtlasCloud now $1.85 per 1M output tokens (was $2.2, -16%)cheapest provider · Jul 27, 2026
  • GLM-4.7: cheapest credible provider via AtlasCloud now $0.52 per 1M input tokens (was $0.6, -13%)cheapest provider · Jul 27, 2026
  • GLM-4.7: cheapest credible provider via Google now $2.2 per 1M output tokens (was $1.85, +19%)cheapest provider · Jul 25, 2026
  • GLM-4.7: cheapest credible provider via Google now $0.6 per 1M input tokens (was $0.52, +15%)cheapest provider · Jul 25, 2026
  • GLM-4.7: context window changed from 202,752 to 204,800 tokenscontext · Jul 21, 2026

Compare GLM-4.7 with

GLM-4.7 vs GPT-5.6 SolGLM-4.7 vs Claude Fable 5GLM-4.7 vs Gemini 3.1 ProGLM-4.7 vs Muse SparkGLM-4.7 vs Grok 4.5

Frequently asked questions

What is GLM-4.7?

GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution.

Who created GLM-4.7?

GLM-4.7 is an AI model developed by Z.ai (Zhipu AI), released Dec 22, 2025.

How much does GLM-4.7 cost?

GLM-4.7's list price from Z.ai (Zhipu AI) is $0.60 per million input tokens and $2.20 per million output tokens.

What is GLM-4.7's context window?

GLM-4.7 supports a context window of up to 205K tokens.

How does GLM-4.7 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, GLM-4.7 scores 89.4% on LiveCodeBench, 95.9% on τ²-bench, 85.6% on MMLU-Pro.

Is GLM-4.7 open source?

GLM-4.7 is released with open weights (Open weights (unrestricted)), so the model can be downloaded and self-hosted.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Jul 30, 2026.