NOUMENOpen app
NOUMEN

Z.ai · Open weights

GLM 4.5V Thinking

Added Nov 22, 2025Open weights

Compare models

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

Loading models…

Use in the app

Thinking-enabled GLM 4.5V that surfaces structured reasoning before its final answer. Great for image-grounded analysis, OCR, charts, and deliberate step-by-step responses.

  • Reasoning
  • Vision
Input
Text, Images
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.60
Output
$1.80

Reference prices as published by the model provider. API example

Specifications

Context window
65.5K
Max output
16.4K
Average output (7 days)
534 tokens
Longer than 50% of 468 models

Side by side

GLM 4.5V Thinking 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 4.5V ThinkingZ.aiThis modelLlama 3.3 70BMetaRemove
ProviderZ.aiMeta
Discovery tierOpen weightsOpen weights
AddedNov 22, 2025Jul 3, 2025
Limits
Context window65.54K128KBest
Max output16.38K16.38K
Average output534 tokens339 tokens
Pricing, USD per 1M tokens
Input$0.60Best$1.75
Output$1.80Best$2.75
Cache read—$1.75
Benchmark indexes
Intelligence Index7.67.7Best
Coding Index—11.9
Math Index73.0Best7.7
Capabilities
ReasoningYesNo
VisionYesNo
Reasoning effort——

Public evaluations

Benchmarks

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

  • GLM 4.5V Thinking
  • Llama 3.3 70B

Intelligence Index

  • GLM 4.5V Thinking7.6
  • Llama 3.3 70B7.7

GLM 4.5V Thinking: Better than 11% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GLM 4.5V Thinking: —
    • Llama 3.3 70B: 11.9

  • Math Index

    Composite of the math evaluations

    • GLM 4.5V Thinking: 73.0Top
    • Llama 3.3 70B: 7.7

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • GLM 4.5V Thinking: 0.0%
    • Llama 3.3 70B: 15.7%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GLM 4.5V Thinking: 6.3%Top
    • Llama 3.3 70B: 3.6%

  • GPQA Diamond

    Graduate-level science questions

    • GLM 4.5V Thinking: 68.4%Top
    • Llama 3.3 70B: 49.8%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GLM 4.5V Thinking: 78.8%Top
    • Llama 3.3 70B: 71.3%

  • IFBench

    Following precise output instructions

    • GLM 4.5V Thinking: 34.2%
    • Llama 3.3 70B: 47.1%Top

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • GLM 4.5V Thinking: 60.4%Top
    • Llama 3.3 70B: 28.8%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GLM 4.5V Thinking: 73.0%Top
    • Llama 3.3 70B: 7.7%

  • AIME

    Competition mathematics

    • GLM 4.5V Thinking: —
    • Llama 3.3 70B: 30.0%

  • MATH-500

    Competition math problems

    • GLM 4.5V Thinking: —
    • Llama 3.3 70B: 77.3%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GLM 4.5V Thinking: 5.3%Top
    • Llama 3.3 70B: 3.0%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GLM 4.5V Thinking: 22.5%
    • Llama 3.3 70B: 26.6%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GLM 4.5V Thinking: 0.0%
    • Llama 3.3 70B: 15.7%Top

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

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-4.5v:thinking",
    "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-4.5v:thinking",
    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.