NOUMENOpen app
NOUMEN

Z.ai · Open weights

GLM 4.7 Original Thinking

Added Dec 22, 2025Open weights

Compare models

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

Loading models…

Use in the app

GLM-4.7 original with extended thinking capabilities for complex reasoning.

  • Reasoning
  • Tools
  • Structured output
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.40
Output
$1.93
Cache read
$0.0801

Reference prices as published by the model provider. API example

Specifications

Context window
200K
Max output
65.5K
Average output (7 days)
1.9K tokens
Longer than 90% of 468 models

Side by side

GLM 4.7 Original 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 4.7 Original ThinkingZ.aiThis modelKimi K2 ThinkingMoonshotRemove
ProviderZ.aiMoonshot
Discovery tierOpen weightsOpen weights
AddedDec 22, 2025Nov 6, 2025
Limits
Context window200K262.14KBest
Max output65.54K98.3KBest
Average output1.9K tokens1.9K tokens
Pricing, USD per 1M tokens
Input$0.40Best$0.60
Output$1.93Best$2.50
Cache read$0.0801Best$0.15
Performance
Throughput31.8 tok/s78.6 tok/sBest
Time to first token8.3 s4.6 sBest
Request success (24h)100.0%Best99.6%
Benchmark indexes
Intelligence Index22.2Best22.0
Coding Index45.3—
Math Index95.0Best94.7
Capabilities
ReasoningYesYes
ToolsYesYes
Structured outputYesYes
Reasoning effort——

Public evaluations

Benchmarks

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

  • GLM 4.7 Original Thinking
  • Kimi K2 Thinking

Intelligence Index

  • GLM 4.7 Original Thinking22.2
  • Kimi K2 Thinking22.0

GLM 4.7 Original Thinking: Better than 53% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • GLM 4.7 Original Thinking: 45.3
    • Kimi K2 Thinking: —

  • Math Index

    Composite of the math evaluations

    • GLM 4.7 Original Thinking: 95.0Top
    • Kimi K2 Thinking: 94.7

Agentic work

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • GLM 4.7 Original Thinking: 1,013 Elo
    • Kimi K2 Thinking: —

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • GLM 4.7 Original Thinking: 71.0%
    • Kimi K2 Thinking: 72.0%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • GLM 4.7 Original Thinking: 27.4%Top
    • Kimi K2 Thinking: 23.8%

  • GPQA Diamond

    Graduate-level science questions

    • GLM 4.7 Original Thinking: 85.9%Top
    • Kimi K2 Thinking: 83.8%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • GLM 4.7 Original Thinking: 85.6%Top
    • Kimi K2 Thinking: 84.8%

  • IFBench

    Following precise output instructions

    • GLM 4.7 Original Thinking: 67.9%
    • Kimi K2 Thinking: 68.1%Top

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • GLM 4.7 Original Thinking: 89.4%Top
    • Kimi K2 Thinking: 85.3%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • GLM 4.7 Original Thinking: 95.0%Top
    • Kimi K2 Thinking: 94.7%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • GLM 4.7 Original Thinking: 31.8%Top
    • Kimi K2 Thinking: 31.1%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • GLM 4.7 Original Thinking: 95.9%Top
    • Kimi K2 Thinking: 93.0%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • GLM 4.7 Original Thinking: 71.0%
    • Kimi K2 Thinking: 72.0%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
31.8 tok/s
Faster than 22% of 254 models
Time to first token
8.3 s
Quicker than 16% of 254 models
Prompt cache hits
43.3%
About $0.26 per 1M input tokens with caching
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 94.1%
  • 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-4.7-original: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-4.7-original: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.