NOUMENOpen app
NOUMEN

Z.ai · Open weights

GLM 4.5

Added Apr 15, 2025Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

GLM-4.5 is a Mixture-of-Experts model built for agent-based applications, with 355B total / 32B active parameters and up to 128k context. It supports reasoning, code generation, and agent workflows.

Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.30
Output
$1.30

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
65.5K
Average output (7 days)
696 tokens
Longer than 62% of 468 models

Side by side

GLM 4.5 compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

GLM 4.5Z.aiThis modelGLM 5.3 ThinkingZ.aiRemove
ProviderZ.aiZ.ai
Discovery tierOpen weightsOpen weights
AddedApr 15, 2025Aug 14, 2026
Limits
Context window128K1.05MBest
Max output65.54K131.07KBest
Average output696 tokens2.3K tokens
Pricing, USD per 1M tokens
Input$0.30Best$0.60
Output$1.30$1.10Best
Cache read—$0.20
Performance
Throughput35.7 tok/s123.3 tok/sBest
Time to first token9.6 s7.7 sBest
Request success (24h)100.0%Best99.7%
Benchmark indexes
Intelligence Index12.844.8Best
Coding Index—74.8
Math Index73.7—
Capabilities
ReasoningNoYes
ToolsNoYes
Structured outputNoYes
Reasoning effort—high, max

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • GLM 4.5
  • GLM 5.3 Thinking

Intelligence Index

  • GLM 4.512.8
  • GLM 5.3 Thinking44.8

GLM 4.5: Better than 29% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GLM 4.5: —
    • GLM 5.3 Thinking: 74.8

  • Math Index

    Composite of the math evaluations

    • GLM 4.5: 73.7
    • GLM 5.3 Thinking: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • GLM 4.5: —
    • GLM 5.3 Thinking: 62.2%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • GLM 4.5: —
    • GLM 5.3 Thinking: 29.4%

  • Harvey LAB-AAv1.1

    Legal tasks passed in full, with no material hallucination

    • GLM 4.5: —
    • GLM 5.3 Thinking: 0.3%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • GLM 4.5: —
    • GLM 5.3 Thinking: 1,505 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GLM 4.5: —
    • GLM 5.3 Thinking: 1,649 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • GLM 4.5: —
    • GLM 5.3 Thinking: 11.2%

  • AA-LCRv1.1

    Reasoning across long documents

    • GLM 4.5: 52.7%
    • GLM 5.3 Thinking: 79.7%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • GLM 4.5: —
    • GLM 5.3 Thinking: 17.2%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GLM 4.5: 13.0%
    • GLM 5.3 Thinking: 42.3%Top

  • GPQA Diamond

    Graduate-level science questions

    • GLM 4.5: 78.2%
    • GLM 5.3 Thinking: 91.7%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GLM 4.5: 83.5%
    • GLM 5.3 Thinking: —

  • IFBench

    Following precise output instructions

    • GLM 4.5: 44.1%
    • GLM 5.3 Thinking: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • GLM 4.5: —
    • GLM 5.3 Thinking: 41.9%

  • SciCode

    Scientific computing in Python

    • GLM 4.5: —
    • GLM 5.3 Thinking: 59.0%

  • LiveCodeBench

    Recent competitive programming problems

    • GLM 4.5: 73.8%
    • GLM 5.3 Thinking: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GLM 4.5: 73.7%
    • GLM 5.3 Thinking: —

  • AIME

    Competition mathematics

    • GLM 4.5: 87.3%
    • GLM 5.3 Thinking: —

  • MATH-500

    Competition math problems

    • GLM 4.5: 97.9%
    • GLM 5.3 Thinking: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GLM 4.5: 22.0%
    • GLM 5.3 Thinking: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GLM 4.5: 43.0%
    • GLM 5.3 Thinking: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GLM 4.5: 52.7%
    • GLM 5.3 Thinking: 79.7%Top

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
35.7 tok/s
Faster than 25% of 254 models
Time to first token
9.6 s
Quicker than 11% of 254 models
Prompt cache hits
45.5%
Request success (24h)
100.0%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 100.0%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "z-ai/glm-4.5",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="z-ai/glm-4.5",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Z.ai

Other Z.ai models, starting with the same family.