NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/GLM 4.7
z-ai logo

GLM 4.7

by z-ai · Released Dec 2025

Open Source
52.4
avg score
Rank #137
Compare
Better than 57% of all models
Context
205K tokens (~102 books)
Input $/1M
$0.60
Output $/1M
$2.20
Type
text
License
Open Source
Benchmarks
29 tested
Data as of
About

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. It demonstrates significant improvements in executing complex agent tasks while...

Tested on 29 benchmarks · BenchGecko score 52.4. Top scores: Chatbot Arena Elo — Overall (1441.5%), Chatbot Arena Elo — Coding (1434.5%), OpenCompass — AIME2025 (95.4%).

Looking for similar performance at lower cost?
DeepSeek V3.1 scores 51.8 (99% as good) at $0.25/1M input · 58% cheaper
Capabilities
coding
58.0
#55 globally
reasoning
50.7
#86 globally
math
51.8
#122 globally
knowledge
48.6
#141 globally
agentic
3.1
#82 globally
general
10.9
#195 globally
language
63.7
#99 globally
Benchmark Scores
Compare All
Tested on 29 benchmarks · Ranked across 8 categories
Score Distribution (all 319 models)
0255075100
▲ You are here
codingCompare coding →
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

83.8·
LiveBench — Coding

Regularly refreshed coding problems that avoid data contamination. New problems added monthly to prevent memorization.

73.1·
LiveBench — Agentic Coding

LiveBench coding tasks that require multi-step reasoning and tool use. Tests planning and execution of complex coding workflows.

41.7·
reasoningCompare reasoning →
LiveBench — Reasoning

Regularly refreshed reasoning problems testing logical deduction, spatial reasoning, and analytical thinking.

59.7·
LiveBench — Data Analysis

Fresh data analysis tasks testing ability to interpret tables, charts, and statistical data.

55.2·
SimpleBench

Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.

37.2·
mathCompare math →
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

95.4·
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

83.3·
LiveBench — Mathematics

Regularly updated math problems that test numerical reasoning, algebra, calculus, and combinatorics.

76.0·
Quick compare:Anthropic logovs Claude Sonnet 4.6DeepSeek logovs DeepSeek V3.2DeepSeek logovs DeepSeek V3.1
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · z-ai GLM 4.7
Dec 2025
52.4
$0.60/M in205Kctx29 benchmarks
Jan 2026
$0.06/M in(-0.54)200Kctx(-5K)1 benchmark
See the full GLM 4.7 family →
Recently Happened
GLM 4.7 pricing increased 50%
Sep 25, 2026
Similar Models
Anthropic logo
Claude Sonnet 4.6
Anthropic
52.3$3.00/1M
DeepSeek logo
DeepSeek V3.2
DeepSeek
51.8$0.28/1M
DeepSeek logo
DeepSeek V3.1
DeepSeek
51.8$0.25/1M
Links
Info
z-aiPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@z-ai
Source Code
GitHubHugging Face
BenchGecko API
glm-4-7
Specifications
  • Typetext
  • Context205K tokens (~102 books)
  • ReleasedDec 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.003
Available On
z-ai logoz-ai$0.60
Categories
codingreasoningmathknowledgeagenticgenerallanguage
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Anthropic logoClaude Sonnet 4.6
52.3
DeepSeek logoDeepSeek V3.2
51.8
DeepSeek logoDeepSeek V3.1
51.8
DeepSeek logoDeepSeek R1 Distill Llama 8B
53.1
GLM 4.7 is an open-source text AI model by z-ai, released in December 2025. It has an average benchmark score of 52.4. Context window: 205K tokens.

Related Models

Claude Sonnet 4.6 · AnthropicDeepSeek V3.2 · DeepSeekDeepSeek V3.1 · DeepSeekDeepSeek R1 Distill Llama 8B · DeepSeekGrok 3 Mini Beta · xAI

Benchmarks

Chatbot Arena Elo — OverallChatbot Arena Elo — CodingOpenCompass — AIME2025OpenCompass — IFEvalOpenCompass — GPQA-Diamond

Related Pages

z-ai · Providerz-ai · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-16

  • GLM 4.7 by z-ai. BenchGecko score 52.4, rank 137 of 319 scored models (normalized average of public benchmark scores).
  • List price $0.60 input · $2.20 output per 1M tokens (as of 2026-10-16).
  • Sold by 7 providers (as of 2026-10-16): DeepInfra (fp4) $0.32 in / $1.40 out · DeepInfra (fp4) $0.40 in / $1.75 out · Venice (fp4) $0.40 in / $1.93 out · Novita (fp8) $0.54 in / $1.98 out · Google $0.60 in / $2.20 out · and 2 more. Every provider

How to cite · data as of 2026-10-16

GLM 4.7 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-16. https://benchgecko.ai/model/glm-4-7

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP